assessment and control of bacillus cereus jaaskelainen 2008
TRANSCRIPT
Assessment and control of Bacillus cereus
emetic toxin in food
Elina Jääskeläinen
Department of Applied Chemistry and Microbiology
Division Microbiology
University of Helsinki
Academic dissertation in Microbiology
To be presented, with the permission of the Faculty of Agriculture and Forestry of the
University of Helsinki, for public criticism in Auditorium 2041 at Viikki Biocenter,
Viikinkaari 5, on February 1th, 2008, at 12 o´clock noon
Helsinki 2008
Supervisor: Prof. Mirja S. Salkinoja-Salonen
Department of Applied Chemistry and Microbiology
Faculty of Agriculture and Forestry
University of Helsinki
Helsinki, Finland
Reviewers: Prof. Willem M. de Vos
1) Laboratory of Microbiology
Agrotechnology and Food Sciences Group
Wageningen University and Research Centre
Wageningen, the Netherlands
2) Department of Basic Veterinary Sciences
Faculty of Veterinary Medicine
University of Helsinki
Helsinki, Finland
Dr. Christophe Nguyen-The
Institute of Plant Products Technology
French National Institute for Agricultural Research (INRA)
University of Avignon
Avignon, France
Opponent: Prof. Jacques Mahillon
Laboratory of Food and Environmental Microbiology
Université Catholique de Louvain
Louvaine-la-Neuve, Belgium
ISNN 1795-7079
ISBN 978-952-10-4458-8 (hardback)
ISBN 978-952-10-4459-5 (PDF)
Yliopistopaino
Helsinki, Finland 2008
Front cover: Boys evaluating Mother`s art of cooking
To my family
Contents:
List of original publications...................................................................................................3
The author´s contribution.......................................................................................................4
Abbreviations.........................................................................................................................5
Abstract..................................................................................................................................6
1. Backround..........................................................................................................................8
2. Review of the literature......................................................................................................8
2.1 The Bacillus cereus group..........................................................................8
2.2 The species Bacillus cereus.......................................................................10
2.2.1 Species description............................................................10
2.2.2 The genome of Bacillus cereus.........................................12
2.2.3 Detection and isolation of Bacillus cereus........................12
2.2.4 Spores of Bacillus cereus..................................................13
2.2.5 Bacillus cereus in the environment and in food................15
2.3 Bacillus cereus food poisoning..................................................................17
2.3.1 Diarrhoeal food-borne infection by B. cereus
and its causative agents..............................................................18
2.3.2 Emetic food-borne intoxication.........................................19
2.3.3 Specific features of emetic toxin-producing
strains of B. cereus.....................................................................23
2.3.4 Methods for detecting and quantifying cereulide..............24
2.4 Emetic toxin production by B. cereus in different growth environments..26
3. Aims of the study...............................................................................................................29
4. Materials and methods.......................................................................................................30
1
5. Results and discussion.........................................................................................................31
5.1 A new method for screening B. cereus isolates for cereulide
production...................................................................................................31
5.2 LC-MS-based quantative analysis of cereulide...........................................32
5.3 Method for direct extraction and analysis of cereulide in foods.................35
5.4 Mining for toxin-producing B. cereus from food........................................36
5.5 Cases of emetic B. cereus food poisoning...................................................39
5.6 The case of acute liver failure......................................................................41
5.7 Analysis of the toxicity target of cereulide in mammalian somatic
and germ cells....................................................................................................42
5.8 Cereulide production under different environmental conditions.................43
5.8.1 Cereulide production by emetic B. cereus
in laboratory cultivation media (cereulide contents
of the harvested bacterial biomass)..............................................43
5.8.2 Time course of cereulide production...................................44
5.8.3 Cereulide production in foods.............................................45
5.8.4 Cereulide production under different atmospheres.............47
6. Conclusions..........................................................................................................................51
7. Tiivistelmä...........................................................................................................................54
8. Acknowledgements..............................................................................................................56
9. References............................................................................................................................59
2
List of original publications:
I. Maria A. Andersson, Elina L. Jääskeläinen, Ranad Shaheen, Tuula Pirhonen, Luc M.
Wijnands, Mirja S. Salkinoja-Salonen. 2004. Sperm bioassay for rapid detection of cereulide-
producing Bacillus cereus in food and related environments. International Journal of Food
Microbiology. 94: 175-183
II. Elina L. Jääskeläinen, Max M. Häggblom, Maria A. Andersson, Liisa Vanne, and Mirja
S. Salkinoja-Salonen. 2003. Potential of Bacillus cereus for producing emetic toxin, cereulide,
in bakery products: quantitative analysis by chemical and biological methods. Journal of Food
Protection. 66: 1047-1054
III. E.L. Jääskeläinen., V. Teplova, M.A. Andersson, L.C. Andersson, P. Tammela, M. C.
Andersson, T.I. Pirhonen, N. –E.L. Saris, P.Vuorela, M.S. Salkinoja- Salonen. 2003. In vitro
assay for human toxicity of cereulide, the emetic mitochondrial toxin produced by food
poisoning Bacillus cereus. Toxicology in Vitro 17: 737-744
IV. E.L. Jääskeläinen, M.M. Häggblom, M.A.Andersson, M.S. Salkinoja-Salonen. 2004.
Atmospheric oxygen and other conditions affecting the production of cereulide by Bacillus
cereus in food. International Journal of Food Microbiology 96: 75-83.
3
The author´s contribution
Paper I. Elina Jääskeläinen was responsible for the experimental work on LC-MS and wrote
the article together with the other authors.
Paper II. Elina Jääskeläinen wrote the article and is the corresponding author. She also
planned and carried out the experimental work except for some of the pH assays and the aw
assays.
Paper III. Elina Jääskeläinen wrote the article and is the corresponding author. She also
planned and carried out the experimental work except for the manual extraction of cereulide,
some of the boar sperm cell assays, cells cultivation and the Paju cell assays.
Paper IV. Elina Jääskeläinen wrote the article and is the corresponding author. She also
planned and carried out all the experimental work.
4
Abbreviations
aw Water activityATCC American Type Culture CollectionbceT Enterotoxin T (Bacillus cereus)Caco-2 cells Colon adenocarcinomaCalu-3 cells Lung adenocarcinomacfu Colony-forming unitCytK Cytotoxin KD95°C Decimal reduction time at 95°CD-value Decimal reduction timeBCET-RPLA Bacillus cereus enterotoxin reversed passive latex agglutinationBHI Brain heart infusionFDA Food and Drug Administration (USA)ESI Electrospray ionizationEU European UnionHBL Haemolytic enterotoxin by B. cereusHeLa cells Derived from cervical cancer cellsHPLC High-pressure liquid chromatographyIDF International Dairy FederationISO International Organization for StandardizationJC-1 5,5´,6,6´-tetrachloro-1,1´,3,3´-tetraethylbenz-
imidazolocarbocyanine iodidekb KilobasepairsLeu LeucineLC-MS Liquid chromatography-mass spectrometryLog KOW Logarithm of the n-octanol -water partition coefficientMb Megabasepairsm/z Mass-to-charge ratioMS Mass spectrometryNCBI National Center for Biotechnology InformationNK cells Natural killer cellsNMKL Nordic Committee on Food AnalysisMYP Mannitol egg yolk polymyxin agarNHE Nonhaemolytic enterotoxinPaju cells Human neural cell linePCR Polymerase chain reactionThr ThreonineTSA Tryptone soy agarTSB Tryptone soy broths.l. Sensu latos.s. Sensu strictoUV UltravioletVal Valine
m Mitochondrial inner membrane transmembrane potentialO-val 2-hydroxyisovaleric acidO-leu 2-hydroxyisocaproic adic
5
Abstract
Despite of improving levels of hygiene, the incidence of registered food borne diseases has
been at the same level for many years: there were 40 to 90 epidemics in which 1000-9000
persons contracted food poisoning through food or drinking water in Finland. Until the year
2004 salmonella and campylobacter were the most common bacterial causes of food borne
diseases, but in years 2005-2006 Bacillus cereus was the most common. Similar
developement has been published i.e. in Germany already in the 1990´s. One reason for this
can be cereulide, the emetic toxin of Bacillus cereus. Bacillus cereus is a common
environmental bacterium that contaminates raw materials of food. Otherwise than salmonella
and campylobacter, Bacillus cereus is a heat resistant bacterium, capable of surviving most
cooking procedures due to the production of highly thermo resistant spores. The food
involved has usually been heat treated and surviving spores are the source of the food
poisoning. The heat treatment induces germination of the spore and the vegetative cells then
produce toxins. This doctoral thesis research focuses on developing methods for assessing and
eliminating risks to food safety by cereulide producing Bacillus cereus. The biochemistry and
physiology of cereulide production was investigated and the results were targeted to offer
tools for minimizing toxin risk in food during the production.
I developed methods for the extraction and quantitative analysis of cereulide directly from
food. A prerequisite for that is knowledge of the chemical and physical properties of the
toxin. Because cereulide is practically insoluble in water, I used organic solvents; methanol,
ethanol and pentane for the extraction. For extraction of bakery products I used high
temperature (100°C) and pressure (103.4 bars). An alternative for effective extraction is to
flood the plain food with ethanol, followed by stationary equilibration at room temperature. I
used this protocol for extracting cereulide from potato puree and penne. Using this extraction
method it is also possible to extract cereulide from liquid food, like milk. These extraction
methods are important improvement steps for studying of Bacillus cereus emetic food
poisonings. Prior my work, cereulide extraction was done using water. As the result, the yield
was poor and variable.
6
To investigate suspected food poisonings, it is important to show actual toxicity of the
incriminated food. Many toxins, but not cereulide, inactivate during food processing like
heating. The next step is to identify toxin by chemical methods. I developed with my
colleague Maria Andersson a rapid assay for the detection of cereulide toxicity, within 5 to 15
minutes. By applying this test it is possible to rapidly detect which food is causing the food
poisoning. The chemical identification of cereulide was achieved using mass spectrometry. I
used cereulide specific molecular ions, m/z (± 0.3) 1153.8 (M+H+), 1171.0 (M+NH4+), 1176.0
(M+Na+) and 1191.7 (M+K+) for reliable identification. I investigated foods to find out their
amenability to accumulate cereulide. Cereulide was formed high amounts (0.3 to 5.5 μg g-1
wet wt) when cereulide producing B. cereus strains were present in beans, rice, rice-pastry
and meat-pastry, stored at non refrigerated temperatures (21-23°C). Rice and meat pastries are
frequently consumed under conditions where no cooled storage is available e.g. picnics and
outdoor events.
Bacillus cereus is a ubiquitous spore former and is therefore difficult to eliminate from foods.
It is therefore important to know which conditions will affect the formation of cereulide in
foods. My research showed that the cereulide content was strongly (10 to 1000 fold
differences in toxin content) affected by the growth environment of the bacterium. Storage of
foods under nitrogen atmosphere (> 99.5 %) prevented the production of cereulide. But when
also carbon dioxide was present, minimizing the oxygen contant (< 1%) did not protect the
food from formation of cereulide in preliminary experiments. Also food supplements affected
cereulide production at least in the laboratory. Adding free amino acids, leucine and valine,
stimulated cereulide production 10 to 20 fold. In peptide bonded form these amino acids are
natural constituents in all proteins. Interestingly, adding peptide bonded leucine and valine
had no significant effect on cereulide production. Free amino acids leucine and valine are
approved food supplements and widely used as flawour modifiers in food technology. My
research showed that these food supplements may increase food poisoning risk even though
they are not toxic themselves.
7
1. Background
The incidence of foodborne disease has increased during recent years (Varnam and Evans,
1991), despite improvement in hygiene. Food- and waterborne diseases are a significant
cause of morbidity and mortality throughout the world. The reasons for this increase may lie
in recent trends in global food production and changes in food technology in the industrialized
countries. Reporting and diagnostic methods have also developed. Bacillus cereus is an
endospore-forming bacterial species and a common cause of food poisoning in many
countries.
Bacillus cereus produces many types of toxins, two of which are most frequently associated
with food poisonings: 1) the thermolabile enterotoxins that are destroyed when food is heated
and 2) the emetic toxin, which is not inactivated by heating of food (Jay et al., 2005;
Granum, 2007). Thermolabile B. cereus enterotoxins contaminating the raw materials of food
are likely to be detoxified by heat, but no method is known for detoxifying the emetic toxin,
cereulide, in food.
2. Review of the literature
2.1 The Bacillus cereus group
The genus Bacillus is a heterogeneous group of Gram-positive, spore-forming rods that
belongs to the low G+C (Guanine+Cytocine) phylum Firmicutes (Holt et al., 1994). Bacillus
subgroup 1 (Bacillus cereus sensu lato group) comprises the species Bacillus anthracis,
Bacillus cereus, Bacillus mycoides, Bacillus thuringiensis, Bacillus pseudomycoides, Bacillus
weihenstephanensis (Granum 2002; Jensen et al., 2003) and the novel pathogen Bacillus
neocereus (van der Zwet et al., 2000, not yet a validly described species) . A close genetic
relation was observed between all B. cereus group members (Helgason et al., 2000). Several
characteristics have been suggested for differentiation of the B. cereus group (Table 1). The
main diagnostic features of B. cereus sensu lato are their ability to hydrolyze lecithin and an
inability to ferment mannitol.
8
Tabl
e 1.
Diff
eren
tatio
n of
mem
bers
of t
he tr
aditi
onal
B. c
ereu
s gro
up (m
odifi
ed fr
om: G
ranu
m 2
007)
Pro
perti
esB
. cer
eus
B. a
nthr
acis
B. t
hurin
gien
sis
B. m
ycoi
des
B. p
seud
omyc
oide
sB
. wei
hens
teph
anen
sis
Mot
ility
+/-
- +
/--
- +
/-Pe
nici
llin s
usce
ptib
ility
- +
- -
n.d
-M
anni
tol f
erm
enta
tion
- -
- -
- -
Cry
stal
line
para
spor
al in
clus
- -
+ -
- -
Hem
olys
is +
- +
(+)
n.d
+
B. w
eihe
nste
phan
ensi
s can
be
dist
ingu
ishe
d fro
mB.
cer
eus
base
d on
gro
wth
at <
7°C
and
not
at 4
3°C
.B. p
seud
omyc
oide
s is
not
dist
ingu
ishab
le fr
omB.
myc
oide
s by
phys
iolo
gica
l or m
orph
olog
ical
cha
ract
erist
ics,
but i
s diff
eren
tiate
d ba
sed
on fa
tty a
cid
com
posit
ion
and
16S
rRN
A g
ene
sequ
ence
s. n.
d. =
not
det
erm
ined
.
9
2.2 The species Bacillus cereus
2.2.1 Species description
Bacillus cereus was first isolated in 1887 from cowshed air by Frankland and Frankland
(Roberts et al., 1996; Forsythe, 2000; Griffiths and Schraft, 2002). The cells under the
microscope are rod-shaped, straight, typically 0.5-2.5 µm in diameter and 1.2-10 µm in length
and are often arranged in pairs or chains (Holt et al., 1994). The cells stain Gram-positive,
although positive staining is often difficult to obtain in older cultures (Varnam and Evans,
1991). The cells are motile by peritrichous flagelli (Varnam and Evans, 1991). B. cereus
endospores are centrally or pericentrally positioned. Figure 1 illustrates the spores of B.
cereus. B. cereus is a mesophilic, facultatively anaerobic bacterium (Griffiths and Schraft,
2002) and is able to grow at redox-potential below -200 mV (Varnam and Evans 1991). B.
cereus has an absolute requirement for three L- amino acids: threonine, leucine, valine (Agata
et al., 1999) as growth factors, but vitamins are not required (Griffiths and Schraft, 2002). The
temperature range of growth is 4 - 55°C (optimum 30-40 °C) (Roberts et al., 1996).
Psychrotrophic strains are common and growth may occur at temperatures of 4-5 °C. The
minimum water activity (aw) for growth is 0.93 and the pH range is 4.3 - 9.3 (Forsythe, 2000).
10
Figure 1. Transmission electron micrograph of a sporulating culture of the emetic B. cereus
strain F4810/72. The culture was grown for 10 days on Tryptic Soy Agar. Bar, 80 nm. Image
taken by Maria Andersson.
-
11
2.2.2 The genome of Bacillus cereus
At least 12 B. cereus strains have been fully genome-sequenced by to date, (National Centre
for Biotechnology Information NCBI, 2007). The published strains with finished sequence are
listed in Table 2. The peptide synthase gene responsible for the nonribosomal synthesis of
cereulide in emetic B. cereus strains (Ehling-Schulz et al., 2005a) is extrachromosomal,
located on an app. 200-kb plasmid (Hoton et al., 2005; Ehling-Schulz et al., 2006; Rasko et
al., 2007). Even through the cereulide synthesis gene is located on a plasmid, the emetic
strains that would produce both cereulide and the haemolytic diarrhoeal toxin, HBL, have not
yet been reported (Guinebretière et al., 2002).
Table 2. The published complete genome sequences of B. cereus strains.Bacillus cereus
strain Size (bp)Number of
genes Links Reference
ATCC10987 5224283 5603 NCBI, Refseq NC003909 Rasko et al., 2004ATCC 14579 5411809 5234 NCBI, Refseq NC004722 Ivanova et al., 2003
ZK 5300915 5134 TIGR, Tax 288681 Han et al., 2006NVH 391-98 4087024 4165 NCBI, Refseq NC009674 http://img.gi.doe.gov
E33L 5300915 5269 NCBI, Refseq NC006274 Han et al., 2006
2.2.3 Detection and isolation of Bacillus cereus
The method for the enumeration of B. cereus in foods has been standardized by the
International Organization for Standardization (ISO, 2004). The method is based on growth
on mannitol egg yolk polymyxin (MYP) agar, in which the polymyxin B serves as a selective
agent to suppress Gram-negative bacteria. The agar base contains D-mannitol as the
fermentation substrate and phenol red as the indicator to detect formation of acid from
mannitol. B. cereus cannot ferment mannitol, and thus no acid will be formed, while the
colonies of B. cereus are pink due to the phenol red. The egg yolk produces a zone of
precipitation around colonies with lecithinase activity, as is the case for most strains of B.
cereus sensu stricto.
12
The lecithinase activity and negative reaction for mannitol fermentation are the most typical
characteristics of B. cereus and also the basis for B. cereus identification according to
Association of Official Analytical Chemists (AOAC, 1995), Nordic Committee on Food
Analysis (NMKL, 1997), Food and Drug Administration (FDA, 1998) and International
Dairy Federation (IDF, 1998). However, some B. cereus strains (mainly emetic), do not show
the typical lecithinase reaction (Pirttijärvi et al., 1999).
2.2.4 Spores of Bacillus cereus
All Bacillus species can form heat-stable endospores (for a recent review, see Henriques and
Moran, 2007). B. cereus spores are an important factor in food-borne illness. The spores have
a D95°C from below 1 min to over 30 min. The spores have no detectable metabolic activity
and can survive in the absence of nutrients for many years. The first event in sporulation is an
unequal division of the cytoplasm, resulting in large and small progeny each with the
complete genome. After a series of morphological changes the mother cell lyses and releases
the spore into the environment. There is no more than one spore per cell (Holt et al., 1994).
The process of spore formation requires about 6 h (Henriques and Moran, 2007).
An endospore is a dormant, tough and non -reproductive structure. The primary function of
endospores is to ensure the survival of the bacterium through periods of environmental stress.
The spores are highly resistant to heat, drying, toxic chemicals, UV radiation, gamma
radiation and other adverse environmental factors. Bacillus spores are among the life forms
most resistant to inactivation, with examples of spores being revived from amber 25-40
million years old or from brine inclusions dated at 250 -million years (Sagripanti et al., 2006;
Henriques and Moran, 2007).
13
B. cereus sensu stricto spores have a more hydrophobic surface than any other Bacillus spp.
spores. Therefore, they adhere to surfaces such as steel and plastics and are difficult to
remove during cleaning (Granum 2002, 2007). Studies have revealed that the spores can
adhere to Caco-2 cells in culture, indicating that they may adhere to the intestinal epithelium
(Granum, 2007). The spores of some B. cereus strains are more resistant to heat than those of
other mesophilic Bacillus spp. such as B. subtilis and B. licheniformis (Carlin et al., 2006). B.
cereus spores are capable of surviving most procedures applied in the cooking of food
(Shinagawa et al., 1996). In collaboration with our laboratory, Carlin et al. (2006)
investigated the heat tolerance of the spores of 17 cereulide-producing strains and 83
cereulide-nonproducing strains of B. cereus and reported that the spores of the emetic strains
were many-fold more heat-resistant than those of the nonproducers. The spores of the strains
producing emetic toxin exhibited higher D-values (P < 0.001) at 90 °C, as well as higher
survival rates after 120 min of heating at 90 °C (P < 0.001), than did those of the non emetic
strains (Carlin et al., 2006). These experimental facts show that emetic B. cereus strains in
food are very difficult to destroy.
Since the spores are metabolically dormant, they must return to active growth, which they do
through the process of germination. Germination consists of a series of degradative events,
during which the various permeability barriers responsible for a significant degree of
endospore resistance properties are broken down. These events result in rehydration of the
core, facilitating entry of molecules from the external environment (Cronin and Wilkinson
2007; Henriques and Moran, 2007). The major germinant of B. cereus spores is inosine
(Yousten, 1975; Hornstra et al., 2007). Glycine and other neutral L-amino acids and purine
ribosides induce germination (Warren and Gould, 1968; Griffiths and Schraft 2002; Hornstra
et al., 2006). L-alanine is the most effective amino acid stimulating germination (Yousten
1975; Griffiths and Schraft 2002; Hornstra et al., 2007).
14
2.2.5 Bacillus cereus in the environment and in food
B. cereus is found in a wide range of habitats (Beattie and Williams, 2000), e.g. in soil and
vegetation. The primary habitat of B. cereus sensu lato is most likely in the gut of arthropod
invertebrates (Jensen et al., 2003; Swiecicka and Mahillon, 2006), but it also colonizes the gut
of small wild mammals including rodents and insectivores (Swiecicka and Mahillon, 2006).
Since this bacterium is widespread in the environment, it enters the food chain through raw
materials. It is a major problem in convenience foods and mass catering (Guinebretière et al.,
2006). The high resistance of the spores allows B. cereus to survive most drying and cooking
processes. The organism grows well in cooked food because of the lack of a competing
microbiota. B. cereus has been isolated from practically all nonsterile foods (Kolst∅ et al.,
2002; Granum, 2007). Few environments have been studied for the presence of cereulide
producers. In the environments studied, only a minority of the B. cereus strains were
cereulide producers (Table 3). In some foods, e.g. beans (Mikami et al., 1994), cereulide-
producing strains may be a substantial group.
15
Tabl
e 3.
Rep
orte
d fre
quen
cies
of e
met
icB.
cer
eus s
train
s in
the e
nviro
nmen
t and
in fo
ods.
Num
ber o
f iso
late
dB
. cer
eus
stra
ins
Met
hod
ofE
nviro
nmen
tC
ount
ryTo
tal
Em
etic
cere
ulid
e de
tect
ion
Ref
eren
ce
Infa
nt fo
ods
Finl
and
100
11B
oar s
perm
test
, LC
-MS
Sha
heen
et a
l ., 2
006
Dai
ry p
rodu
ctio
n ch
ain
Swed
en56
6878
Boa
r spe
rm te
st, L
C-M
SSv
enss
onet
al .,
200
6V
ario
us fo
ods
(a)
NL
796
65H
Ep-
2 ce
ll te
stW
ijnan
dset
al .,
200
6V
ario
us fo
ods
(b)
Japa
n31
016
HE
p-2
cell
test
Mik
amie
t al.
, 199
4R
eady
-to-e
at fo
odD
enm
ark
401
PCR
Ros
enqu
iste
t al .,
200
5S
oils
and
ani
mal
faec
esU
K10
10
MTT
-test
Alta
yar a
nd S
uthe
rland
200
5W
ashe
d po
tato
UK
83
MTT
-test
Alta
yar a
nd S
uthe
rland
200
5Fo
od, h
uman
faec
es, e
nviro
nmen
ts (c
)Ja
pan
4338
HE
p-2
cell
test
Nis
hika
wa
et a
l ., 1
996
Food
, hum
an fa
eces
, env
ironm
ents
(d)
Japa
n76
4H
Ep-
2 ce
ll te
stN
ishi
kaw
aet
al .,
199
6P
asta
food
(e)
Finl
and
122
83B
oar s
perm
test
, LC
-MS
Pirh
onen
et a
l ., 2
005
a) o
ils a
nd fa
ts a
nd th
eir p
rodu
cts,
fish
and
mea
t and
thei
r pro
duct
s, m
ilk a
nd m
ilk p
rodu
cts,
pastr
y, v
eget
able
s and
thei
r pro
duct
s, re
ady-
to-e
at fo
ods,
flavo
urin
gsb)
veg
etab
les,
frui
ts, g
rain
, fer
men
ted
food
sc)
isol
ated
from
five
em
etic
-syn
drom
e ou
tbre
aks
d) is
olat
es a
ssoc
iate
d in
oth
er th
an e
met
icB.
cer
eus f
ood
pois
onin
g ou
tbre
aks
e) is
olat
es a
ssoc
iate
d in
with
an
emet
ic-s
yndr
ome
food
-bor
ne o
utbr
eak
16
2.3 Bacillus cereus food poisoning
In European legislation B. cereus is classified as a Hazard group 2 organism, based on its
ability to cause infections in humans (European Commission, 1993). In addition, it is the
causative agent of two distinct types of food poisonings. B. cereus was first proven to cause
food-borne disease in 1950. The food was highly contaminated vanilla sauce and
consumption resulted in a diarrhoeal illness (Jay et al., 2005). About 20 years later B. cereus
was also recognized to cause an emetic type of gastrointestinal disease; in 1971 many cases
associated with B.cereus in fried rice from Chinese restaurants (Mortimer and McCann,
1974). Subsequently, B. cereus was recognized as an important cause of food poisoning
worldwide.
In 2005 a total of 55 food poisoning outbreaks, in which food or drinking water was shown to
be the causative agent, were registered in Finland (Niskanen et al., 2006). The causative agent
for the outbreaks remained unidentified in 19 outbreaks (38%). Five epidemics (10%,
involving 64 persons) were caused by B. cereus, i.e. more than any other recognized bacterial
agent in Finland. In year 2006 similar developement has continue (Niskanen et al., 2007). B.
cereus is also a major problem in convenience foods and mass catering in other European
countries (Guinebretière et al; 2002, 2006; Wijnands et al., 2006). The B. cereus toxins that
cause fatal poisonings in humans include cytotoxin K (Lund et al., 2000) and cereulide
(Mahler et al., 1997; Dierick et al., 2005). In addition, B. cereus may have been involved in
outbreaks involving heated foods, in which no viable bacteria could be isolated. Cytotoxin K
is a protein consisting of a single polypeptide chain (Fagerlund et al., 2004). Cereulide, the
emetic toxin of B. cereus, is a nonribosomally synthesized small peptide that can survive
heating, but neither authorities nor food manufacturers analyse foods or raw materials
routinely for cereulide. The reporting rate of illness caused by B. cereus may also be
underestimated, due to the usually short duration (often < 24 h) of the diarrhoeal and emetic
syndromes (Granum 2007). Consequently the full extent of B. cereus food poisoning in
Finland, as well as in other countries, is yet unknown.
17
The reported food-borne outbreaks and cases attributed to B. cereus in North America,
Europe and Japan range from 1% to 22% for outbreaks covering 0.7- 33% of the cases
(Griffiths and Schraft 2002). The Netherlands and Norway were reported to have the most
extensive problem due to B. cereus (Griffiths and Schraft 2002). The type of food-borne
illness caused by B. cereus varies among countries. In Japan, the emetic syndrome is about
10 times more frequently reported than the diarrhoeal disease, while in Europe the diarrhoeal
illness is more frequently reported. This difference is presumably due to the differences in
food and cooking traditions among these areas (Granum, 2007).
2.3.1 Diarrhoeal food-borne infection by B. cereus and its causative agents
Diarrhoeal illness due to ingestion of B. cereus spores is characterized by abdominal pain and
diarrhoea. The incubation period is 8-16 h and the symptoms persist for 12-24 h (Sim 1998;
Beattie and Williams, 2000; Granum 2007). The diarrhoeal syndrome caused by B. cereus is
mediated by one or the three diarrhoeal enterotoxins (Table 4): the tripartite toxins
haemolysin BL (HBL) and non-haemolytic enterotoxin (NHE), the two forms of cytotoxin K
(cytK-1 and cytK-2) (Fagerlund et al., 2004) and possibly enterotoxin T and enterotoxin FM
(Guinebretière et al., 2002; Moravek et al., 2006) . The proteolytic enzymes and pH of the
gastrointestinal tract digest these enterotoxins if they are preformed in foods. B. cereus spores
survive digestion and may germinate in the intestine (Jensen et al., 2003; Swiecicka et al.,
2006), while the vegetative cells may produce toxin in the gut.
Two immunological assays are commercially available for the detection of B. cereus
diarrhoeal toxins. BCET-RPLA: Bacillus cereus enterotoxin reverse passive latex
agglutination (Oxoid, Basingstoke, UK) detects the L2 component of the haemolysin (Granum
2002). The Tecra (Batley, UK) Bacillus diarrhoeal enterotoxin visual immunoassay (BDE-
VIA) detects the 45- kDA protein of the non-haemolytic enterotoxin (Lund and Granum,
1996). A number of cell lines are also susceptible to the diarrhoeal toxins, e.g. Vero (monkey
kidney) and CHO (Chinese hamster ovary). The diarrhoeal enterotoxin is produced in the gut
by germinating spores of B. cereus (Granum, 2007). Some B. cereus strains may stably
colonize the gut of at least arthropods (Swiecicka and Mahillon, 2006). There is no
documentation on the illness-causing effect of the toxins in ingested food. Most likely the
18
ingested toxin proteins are inactivated in the human digestive tract by proteolytic enzymes.
The presence of diarrhoeal toxin genes can be detected by using polymerase chain reaction
(PCR) (Guinebretière et al., 2002, 2006; Abriouel et al., 2007; Fagerlund et al., 2007). In
food-borne isolates of B. cereus, the presence of a toxin gene does not prove that the
bacterium actually produced the diarrhoeal toxin in the human gut. Since B. cereus in heated
foods is always present as spores, the actual illness will only occur if the spores germinate in
the gut. Wijnands et al. (2007) showed that germinants from differentiated Caco-2 cells
induced spore germination in B. cereus.
Table 4. Enterotoxins known to be produced by Bacillus cereus. Modified from: Granum2002.
Food CommercialToxin Type/ size poisoning detection methodHaemolysin BL Protein, 3 components (46, 38, 37 kDa) Probably Oxoid assayNonhaemolytic Enterotoxin (NHE) Protein, 3 components (41, 40, 36 kDa) Yes Tecra kitCytotoxin K1, K2 Protein, 1 components (34 kDa) Yes NoEnterotoxin T Protein, 1 component (41 kDa) Unknown NoEnterotoxin FM Protein, 1 components (45 kDa) Unknown No
2.3.2 Emetic food-borne intoxication
The causative agent of B. cereus emetic food poisoning is a ring-structured
dodecadepsipeptide 1.2 kDa in size, first identified by Agata et al. (1994). Cereulide consists
of only three repeats of 2 amino acids and 2 hydroxy acids: D-O-leu-D-Ala-L-O-Val-L-Val
(Figure 2). Cereulide structurally resembles valinomycin produced by strains of Streptomyces
(Agata et al., 1994). Both cereulide and valinomycin are potassium ionophores (Mikkola et
al., 1999; Teplova et al., 2006; Andersson et al., 2007). Cereulide is resistant to heat,
extremes of pH and to the proteolytic activities of pepsin and trypsin (Kramer and Gilbert,
1989). If the ingested food contains cereulide, the toxin is likely to remain intact and will
likely become sorbed from the gut in its active toxic form. The molecular properties of
cereulide are listed in Table 5.
19
Together with the cytotoxin K (Lund et al., 2000), cereulide is regarded as the most
dangerous to human health of the toxins produced by B. cereus, because it is responsible for
deaths of young healthy persons. A 17-year-old boy in Switzerland died of fulminant liver
failure caused by mitochondrial damage after consuming food contaminated with B. cereus
and its emetic toxin (Mahler et al., 1997). Similarly, a 7-year-old girl in Belgium died only
13 h after ingesting B. cereus- contaminated pasta salad (Dierick et al., 2005). The
significance of cereulide has probably not been recognized in liver failures of unknown
aetiology.
The toxin preformed in food may cause symptoms 0.5-5 h after ingestion of the contaminated
food. The illness is characterized by nausea and vomiting lasting for 6-24 h. In the stomach,
cereulide will interfere with 5-HT3 (serotonin) receptors of the nervus afferent that enervates
the stomach. Dissecting this nerve resulted in loss of the emetic response to ingested cereulide
in an insectivore, the house musk shrew (Suncus murinus) (Agata et al., 1995). Cereulide
inhibits the cytotoxic activities and cytokine production of human natural killer cells and is
thereby a potential immunosuppressant (Paananen et al., 2002). The toxic properties of
cereulide are listed in Table 6.
Extracts from the emetic strains of B. cereus induced emesis in rhesus monkeys (Macaca
mulatta) and in musk shrews (Suncus murinus) (Table 5). Yokoyama et al. (1999) showed
that mice were insensitive to orally given synthetic cereulide, but when cereulide was injected
intraperitoneally it caused vacuolization and mitochondrial swelling in the liver, similar to
that reported in a fatal human case (Mahler et al., 1997). The hepatocytes showed
mitochondrial swelling, with loss of cristae from the mitochondria, and dose-dependent
increases in small fatty droplets in the cytoplasm. At higher doses of cereulide, massive
degeneration of hepatocytes occurred in the mouse. In the mouse, regeneration of hepatocytes
was observed 4 weeks after exposure to 10 µg of cereulide per mouse, at 25 µg of cereulide
per mouse, the mice died within hours.
Emetic food poisoning is often associated with rice foods. Many incidents of emetic illness
are associated with starchy foods such as mashed potatoes (Jay et al., 2005). Griffiths and
Schraft (2002) suggested that starch may promote the growth of B. cereus and the production
of emetic toxin. Emetic B. cereus strains are mainly unable to hydrolyse starch (Shinagawa
1993; Agata et al. 1996; Pirttijärvi et al., 1999, 2000) and starchy foods will look and taste
20
fine even though emetic B. cereus colony counts may be high. This may explain why emetic
food poisonings are usually associated with starchy foods. The only studies so far, in which
the contents of cereulide were estimated in more than one food implicated in emetic-type B.
cereus food poisoning, were published by Agata et al. (1999, 2002). (Table 7). The cereulide
levels of foods estimated in their study ranged from < 5 to 1280 ng/g. The emetic toxin
content of the foods in their study was estimated, based on the toxicity titres of aqueous
supernatants of foods measured using human larynx carcinoma (HEp-2) cells. Therefore the
numbers are only accurate up to the dilution step. The exact toxin dose in humans is difficult
to measure, even if a more accurate analysis is used, because the toxin is likely to be
inhomogenously distributed in most foods.
Figure 2. Structure of cereulide. Picture from:http://www.biocenter.helsinki.fi/groups/salkinoja/index.htm
21
Tabl
e 5.
Mol
ecul
ar p
rope
rties
of c
ereu
lide.
Ref
eren
ceM
olec
ular
stru
ctur
eC
yclic
dod
ecad
epsi
pept
ide,
115
2g/m
olA
gata
et a
l., 1
994;
Mik
kola
et a
l., 1
999
Synt
hesi
sN
onrib
osom
alH
oton
et a
l. 2
005;
Ehl
ing-
Schu
lzet
al.
, 200
6; R
asko
et a
l., 2
007
Sens
ory
prop
ertie
sC
olou
rless
, odo
urle
ssO
ctan
ol-w
ater
coe
ffici
ent
Log
Kow
6.0
Tepl
ova
et a
l., 2
006
Hea
t sta
bilit
yN
o lo
ss o
f act
ivity
upo
n co
okin
g or
aut
ocla
ving
Mik
ami e
t al.
, 199
4; S
hina
gaw
aet
al.
, 199
6pH
stab
ility
Stab
le b
etw
een
2-11
Mik
ami e
t al.
, 199
4; S
hina
gaw
aet
al.
, 199
6B
lack
-lipi
d m
embr
ane
Pota
ssiu
m io
noph
ore
Mik
kola
et a
l., 1
999;
Tep
lova
et a
l., 2
006;
And
erss
on e
t al.,
200
7Is
olat
ed ra
t liv
er m
itoch
ondr
iaC
atal
yses
affl
ux o
f K+
ions
Tepl
ova
et a
l., 2
006
Mod
e of
cel
l act
ion
Dep
olar
ized
mito
chon
dria
l mem
bran
eof
boa
r spe
rm c
ells
Hoo
rnst
raet
al.
, 200
3of
NK
cel
lsPa
anan
enet
al.,
200
2of
neu
ral c
ells
Tepl
ova
et a
l., 2
004
of is
olat
ed ra
t liv
er m
itoch
ondr
iaK
awam
ura-
Sato
et a
l., 2
005
Tabl
e 6.
Foo
d po
isoni
ng p
rope
rties
of c
ereu
lide,
the
emet
ic to
xin
ofB.
cer
eus.
Ref
eren
ceE
met
ic d
ose
10 µ
g to
xin/
kg r
hesu
s m
onke
y (M
acac
a m
ulat
ta)
Shi
naga
wa
et a
l.,19
95 8
-10
µg to
xin/
kg h
ouse
mus
k sh
rew
(S
uncu
s m
urin
us)
Aga
taet
al.,
199
5In
cuba
tion
perio
d0.
5-6
hB
eatti
e an
d W
illia
ms,
200
0D
urat
ion
of il
lnes
s6-
24 h
Grif
fiths
and
Sch
raft,
200
2P
rodu
ctio
nP
refo
rmed
in fo
odG
ranu
m, 2
007
Tryp
sin
dige
stio
nN
ot c
leav
ed b
y try
psin
Mik
amie
t al.
, 199
4, S
hina
gaw
aet
al.
, 199
6P
reva
lenc
e in
food
sM
any
food
s lik
e, e
.g. r
ice,
pas
tries
, pas
ta, n
oodl
esJa
yet
al.
, 200
5
22
Table 7. Emetic toxin contents in food samples implicated in B. cereus emetic-type foodpoisoning (Agata et al., 2002). Similar foods and emetic toxin contents were published earlierby the same author (Agata et al., 1999).The toxin contents were measured with by HEp-2 cellvacuolation activity of the centrifuged and then autoclaved supernatants of foods. A foodhomogenates were prepared in distilled water, using the stomacher instrument.
Food Cereulide titer (ng/g)Fried rice 1 1280Fried rice 2 160Fried rice 3 160Fried rice 4 < 5Boiled rice 1 640Boiled rice 2 320Boiled rice 3 160Boiled rice 4 80Boiled rice 5 10Spaghetti 1 80Spaghetti 2 40Noodle 20Curry and rice 80
2.3.3 Specific features of emetic toxin-producing strains of B. cereus
Shinagawa (1993), Agata et al. (1996) and Nishikawa et al. (1996) concluded, based on
phenotypic properties, that emetic toxin production was associated with a specific class of
Bacillus. This was later supported by analysis of chemotaxonomic and genotypic properties
(Pirttijärvi et al., 1999; Ehling-Schulz et al., 2005b). In the past decade, more information
accumulated and it is now evident that cereulide-producing strains may be more diverse than
previously believed (Apetroaie et al. 2005; Vassileva et al., 2007). Most currently described
emetic strains of B. cereus share the originally described features, such as being negative for
salicin fermentation and for starch hydrolysis (Shinagawa 1993; Agata et al. 1996), but
exceptions are also being found. Thorsen et al. (2006) recently described two starch-positive
cereulide-producing strains of B. weihenstephanensis. Most of the known cereulide-
producing strains belonged to serovar H1. However, some emetic toxin-producing strains
belong serotypes H3 and H12 (Taylor and Gilbert, 1975; Hughes et al., 1988; Agata et al.,
1996, Vassileva et al., 2007).
23
The B. cereus group strains display an extremely large array of ribopatterns. A majority of the
emetic toxin-producing strains, studied by the year 2000, possessed identical ribopatterns
(Pirttijärvi et al., 1999; Ph. D thesis of Tuija Pirttijärvi 2000). Recently, emetic strains with
novel ribopatterns were reported (Apetroaie et al., 2005; Shaheen et al., 2006).
2.3.4 Methods for detecting and quantifying cereulide
Reagents or equipment for detecting and/or quantifying cereulide are not yet commercially
available. The gold standard for emetic toxin detection is the monkey-feeding assay (Griffiths
and Schraft, 2002). However, by European legislation (Registration, Evaluation,
Authorization and Restriction of Chemicals, REACH), effective as of June 1st 2007, whole
animals are not allowed for food testing. Consequently, in vitro assays must be used for toxin
detection in food. Cereulide causes vacuolation of the mitochondria in HEp-2 (human larynx
carcinoma) cells. This can be visualized under a light microscope (Hughes et al., 1988).
Finlay et al. (1999) described a modified, more sensitive HEp-2 cell test. The method is based
on the use of 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT), which
detects the mitochondrial dehydrogenase, regarded as an indicator of cell viability. The
detection limits of various cereulide assays are compiled in Table 8.
Our laboratory developed a test (Andersson et al., 1998) based on boar spermatozoan
motility. The plasma membrane of boar sperm cells has a low sterol content and is therefore
highly permeable to hydrophobic molecules, such as cereulide (Table 5). The motility of boar
spermatozoa is dependent on correct functioning of the mitochondria. Inhibited motility may
be an indication of mitochondrial damage. Motility inhibition can be observed using a light
microscope (Andersson et al., 1998) or using a commercially available sperm analyser
(Rajkovic et al., 2006b). The mitochondrial electric transmembrane potential (Δψm) of
mammalian cells can be visualized by staining with JC-1 (5, 5’, 6, 6’-tetrachloro-1, 1’, 3, 3’-
tetraethylbenzimidazolylcarbocyanine iodide). The lipophilic fluorochrome JC-1 changes its
emission spectrum, depending on the level of Δψ (Reers et al., 1995). The first effect visible
after exposure of sperm cells to cereulide was hyperpolarization of the plasma membrane,
which occurred within 5 min of exposure to the bacterial extract. Subsequently, the sperm
cells lost motility and the mitochondria became depolarized.
24
Cereulide easily dissolves in organic solvents and can thus be identified and quantitated by
high-pressure liquid chromatography (HPLC) combined with mass spectrometry (MS). The
first report for this accurate cereulide quantification method was published in 2002
(Häggblom et al., 2002). Separation was done, using HPLC and quantitative analysis by
determining the concentration of the indicative mass ions specific for cereulide with an ion
trap mass spectrometer. Cereulide and valinomycin have similar responses (same toxin
concentration → same peak area in MS) in HPLC-MS analysis (Häggblom et al., 2002).
Since purified cereulide is not commercially available for use as a standard, valinomycin was
used as a reference for quantification of cereulide.
PCR-based assays were recently developed for identifying potentially emetic B. cereus
strains (Ehling-Schulz et al. 2004, Horwood et al., 2004). The PCR is a reliable method for
detecting the presence of the cereulide synthase gene (Fricker et al., 2007) . The presence of
the toxin synthase gene is required for producing cereulide. However, its presence does not
show whether the bacterium actually produces or if the food contains cereulide in
concentrations sufficient to cause disease.
The actual production of cereulide is strain-dependent (Apetroaie et al., 2005) and also
strongly affected by the environment (Shaheen et al., 2006) . For assessing the risk of food
poisoning by emetic B. cereus, direct analysis of this toxin in food is needed. Risk assessment
should therefore be based on the toxin actually present in the food or the likelihood of toxin
formation in food or in the gut.
Table 8. Lowest limit of various methods used for cereulide detection.
Method for detection Toxicity endpoint Detection limit ReferenceToxicity titer based methodsHEp-2 cells Vacuolation ca. 1-5 ng per ml Mikami et al. , 1994MTT method Vacuolation, staining ca. ≥ 0.5 ng per ml Finlay et al., 1999Boar sperm cell test Motility 0.5 ng per ml Andersson et al., 1998Direct chemical analysisLC-MS 10 pg per injection Häggblom et al., 2002
25
2.4 Emetic toxin production by B. cereus in different growth environments
Various growth media have been evaluated as production media for the emetic toxin of B.
cereus. Szabo et al. (1991) first reported that commercial skim milk is a good environment
for the B. cereus heat-stable toxin production measured by the HEp-2 cell assay. Cereulide is
so far the only known heat-stable B. cereus toxin found in foods. Therefore, Szabo et al.
(1991), and many other authors (Table 6 and Agata et al., 1999, 2002) who used toxicity
assays to measure the emetic B. cereus toxin most likely measured toxicity caused by
cereulide. Molecular identification of cereulide was done later by Agata et al. (1994) and by
Mikkola et al. (1999). Wang et al. (1995) described homocereulide with a molecular mass of
1166 Da. These authors isolated homocereulide from a marine B. cereus strain SCRC and
showed its potent cytotoxicity. However, homocereulide has never been shown to act as the
emetic toxin nor has it been associated with food poisonings.
Agata et al. (1999) reported that cereulide titres were higher in commercial skim milk media
than in brain heart infusion broth (BHI), trypticase soy broth (TSB) or nutrient broth when
these were preinoculated with the same strain of B. cereus, NC7401. These authors also
developed a chemically defined medium for cereulide production (Agata et al., 1999). Three
amino acids: L-valine, L-leucine and L-threonine are essential for B. cereus growth as well as
for the production of cereulide.
Agata and coworkers (2002) studied cereulide production in different foods. Various
consumer foods were seeded with B. cereus strain NC7401, added in amounts of 103 cfu/g.
After incubation for 24 h, they prepared the food as a suspension in distilled water, cleared the
suspension by centrifugation and then measured in the supernatant the titer of heat-stable
toxicity compared with those of other food tested. They obtained (based on toxicity) the
highest cereulide contents (320 ng/g) in boiled rice. Water extracts from similarly treated
bread and cake contained only 20 ng of cereulide per g of extracted food when the incubation
time and temperature were the same. In egg and its products only low amounts, < 5-10 ng /g,
of cereulide were extracted.
Szabo et al. (1991) also found that white rice, inoculated with B. cereus strain F4810/72,
accumulated at 27 °C within 18 h more water-extractable heat-stable toxin (toxin titer 512)
26
than did brown rice (toxin titer 128) or converted rice (toxin titer 256). In the present study
the toxicity assay was not calibrated, so the toxin titers cannot be compared with those in
other works.
The cereulide productivity of B. cereus strains was reported to be sensitive to ambient
temperatures (Häggblom et al., 2002). Szabo et al. (1991) reported the optimum temperature
for emetic toxin production as 20 - 30 °C. Most emetic strains of B. cereus grow at
temperatures of over 40 °C (Häggblom et al, 2002) and some up to 52 °C (Carlin et al., 2006;
Ehling-Schulz et al., 2006). Cereulide production by the strain F4810/72 was nondetectable at
8 °C and at 40 °C (Häggblom et al., 2002).
Finlay et al. (2000) showed that low temperatures (10 °C) suppressed the growth of and thus
also emetic toxin production by the B. cereus strains F4810/72, F3748/75, F3744/75,
F4562/75, F4552/75, F2427/75 and F2549A , F5881, F4810/72, NS117 and NS115 in skim
milk medium. Similarly, Häggblom et al. (2002) showed that cereulide production by B.
cereus strains F4810772, NC7401 and F5581 was detectable, but low below 12 °C in tryptic
soy broth. Rajkovic et al. (2006a), used brain hearth infusion broth as the medium and
showed that B. cereus strains F4810/72, NS115 and NS117 produced no emetic toxin at
12 °C. Thorsen et al. (2006) reported that two strains of the psychrotolerant species, B.
weihenstephanensis, may produce cereulide. These two strains grew at temperatures as low as
8 °C, but produced cereulide only at 25 °C.
Häggblom et al. (2002) reported that cereulide production by the B. cereus strains NC7401
and F4810/72 in stationary incubated Trypticase soy broth was undetectable ( < 0.02 µg ml-1)
compared with cultures incubated on a rotary shaker at 150 rpm ( > 1 µg ml-1) during 24 h .
Agata et al. (2002) and Finlay et al. (2002) observed 90% more emetic toxin production in
shaken milk as than in stationary incubated milk. However, Shaheen et al. (2006) inoculated
infant food formulas with the B. cereus strain F4810/72. They found little cereulide in dairy-
based formulas, whether shaken or not, but found much higher cereulide concentrations (50 x)
when cereal-based infant formula foods were stationary-incubated compared with moderate
(60 rpm) shaking for 24 h at 21-23 °C.
27
Rajkovic et al. (2006a) studied cereulide production in laboratory media under atmospheres
with differing oxygen contents. In their study the head space gas composition was controlled
with a CO2/O2 gas analyser. These authors found that no cereulide accumulated in TSA plate-
grown cultures (B. cereus emetic strains NS117 and 5964a) when the atmosphere contained
less than 1.6 vol % O2, but when the O2 concentration was 4.5 vol %, high amounts (about
1000 ng mg-1) of cereulide accumulated.
28
3. Aims of this study
This doctoral thesis research focuses on developing methods for assessing and eliminating
risks to food safety by cereulide-producing Bacillus cereus. The biochemistry and physiology
of cereulide production were investigated and the results targetted to offer tools for food
production technology to minimize the toxin risk.
The specific goals were to:
1. Develop methods useful for rapid scoring of cereulide production among B. cereus isolates
from foods or from the environment.
2. Develop methods for quantitative extraction and analysis of cereulide directly from food.
3. Identify conditions under which cereulide production by B. cereus may occur or not occur
in selected growth media or foods.
4. Compare the mitochondrial toxic effects of cereulide in mammalian somatic cells and germ
cells.
29
4. Materials and methods
The methods used in this study are listed in Table 9.
Table 9. Methods used in this studyAnalysis Description Reference, manufacture
Extraction methods:Extraction of cereulide from Paper Ibacterial cultures
Extraction of cereulide from Paper IIfood
Extraction of cereulide from Chapter 5.6liver
Assays for toxicity:Boar sperm motility inhibition Paper I Andersson et al.,1998
Bull sperm motility inhibition Paper III
Caco2 (colon carcinoma) cell Paper IIIexposure to cereulide
HeLa (cervical cancer) cell Paper IIIexposure to cereulide
Paju (human neuroblastoma) Paper IIIexposure to cereulide
Calu-3 (human lung carcinoma) Paper IIIexposure to cereulide
JC-1 staining for detecting Paper III Reers et al., 1995electric transmembranepotentials in cells
Chemical methods for cereulide:LC-MS of cereulide Paper II Häggblom et al., 2002
Methods for characterizationof B. cereus
Haemolysis Paper I
Bacillus cereus enterotoxin Chapter 5.4 Beecher and Wong,1994BCET-RPLA
Anaerobic incubation Paper IVin ≥ 99.5% N2
Anaerobic incubation in Chapter 5.8.4 Oxoid (Cambridge, UK)CO2 9-13%, O2 < 1% anaerobic bags (Code:
AN0035)Indicator code BR0055
30
5. Results and discussion
5.1 A new method for screening B. cereus isolates for cereulide production
In the present thesis I describe a novel rapid bioassay for detection of cereulide. The method
is based on the inhibition of sperm motility within 5 min of exposure (Paper I). The test may
be carried out with a single colony from the primary isolation plate with no need to prepare
pure cultures for the toxicity assay. The toxicity threshold for the boar spermatozoa in this
rapid assay was 2 ng of cereulide per ml. Boar semen for artificial insemination is
commercially produced through out the year and therefore readily available.
Steps in the rapid boar sperm microassay
1. A loopful (about 10 mg wet wt) of biomass is picked from a single colony on
an agar plate (e.g. 28 °C, 20-24 h) and suspended in 200 μl of methanol
in a capped tube (about 4 ml)
2. The tube is capped and placed in boiling water for 15 min.
3. The tube is cooled and then vortexed for 2 min.
4. Aliquots of 0.5 - 10 μl of the obtained suspension are dispensed into 200 μl
of extended boar semen and incubated in a thermoblock at 37 °C.
5. After 5 min of exposure, the motility of the sperm cells is recorded
visually, using a phase-contrast microscope.
To date cereulide is the only heat-stable food poisoning toxin known to inhibit sperm motility
within an exposure time of only 5 min. The other toxins of B. cereus inhibit sperm motility
only when the exposure time is much longer. This fact explains why the outcome of the rapid
boar sperm microassay is specific to cereulide.
We used the rapid sperm microassay to search for toxic B. cereus strains in a pasta food
incriminated in a food poisoning case (Paper III). A toxin positive result (= inhibited sperm
motility) was obtained for 83 strains of the 122 tested. Two researchers prepared the extracts
and executed the microscopic analyses during one working day. In the earlier version of the
boar sperm test (Andersson et al., 1998), the same task would have required much longer
periods of time (Table 10).
31
Later, Rajkovic et al. (2006b) used a variant of our method, based on computer-aided boar
semen motility analysis (Hamilton Thorne Ceros 12.1, Hamilton Thorne Biosciences,
Beverly, MA, USA) for cereulide detection. In their protocol, the calibration curve was
limited to the concentration range of 20-400 ng of cereulide per ml of the extracted bacteria or
food. The boar sperm test is also used in Norway at the laboratory of P. E. Granum
(Department of Food Safety and Infection Biology, Norwegian School of Veterinary Science,
Oslo) which serves as the national reference laboratory for B. cereus (From et al., 2007).
Table 10. The originally described sperm assay (Andersson et al., 1998) and the rapid sperm
assay (Paper I): comparison of the essential features.
Sperm assay Rapid sperm microassay(Andersson et al ., 1998) in this thesis
Culture time (20-30 °C) 10 days 1 dayBiomass needed (wet wt) 100 mg 5-10 mgTime to make one extract ca. 10 h 15-30 minExposure time 1-4 days 5-15 minDetection limit 1 ng cereulide per test 0.4 ng cereulide per test
5.2 LC-MS-based quantative analysis of cereulide
The bioassays measure the toxic effects and not the toxic substances, but the quantification of
toxin requires chemical methods. In my thesis, I executed a conclusive analysis of cereulide,
using LC-MS (Papers I, II). HPLC was used for separation and detection was performed
with ion trap MS. The column used was a Discovery C-8, 100 mm × 2.1 mm and 5-µm
particle size (Sigma-Aldrich Corp., St. Louis, Mo, USA). The mobile phase consisted of 95%
acetonitrile with 4.9% water with 0.1% trifluoroacetic acid at a flow rate of 0.15 ml min-1 at
25 °C. The effluent from the HPLC was fed into an electrospray ionization (ESI) ion trap
mass analyser. My protocol for quantification differed from that described by Häggblom et al.
(2002), in that I used specific ions rather than integration of the total chromatogram over a
mass range of 500-1300 m/z. I used ion ranges (m/z) of 1153-1155, 1170-1172, 1175-1177
and 1191-1193. These are specific for the molecular adducts of cereulide with [H+], [NH4+],
[Na+] and [K+], respectively.
Using the above method, we analysed hundreds of specimens for their cereulide contents
(examples shown in Table 11). The specimens represented a wide range of geographic and
material origins. Table 11 shows that the toxicity titer measured with the rapid boar sperm
32
microassay corresponded very well with the actual cereulide concentration measured with
LC-MS. As is seen from Table 11, the cereulide content of the bacterial biomass varied
widely among the strains even when grown under identical conditions. Low producer strains
contained only a few nanograms of cereulide per mg of B. cereus biomass (wet wt), while
other strains produced up to 1000 times more cereulide. The method developed by me (this
thesis) was subsequently used in other studies in our laboratory and elsewhere. The studies
also showed wide differences in cereulide productivity among strains (Apetroaie et al., 2005;
Carlin et al., 2006; Shaheen et al., 2006). The reasons for the different productivities are as
yet unknown, but they show that among the B. cereus emetic strains there are genetic and
physiological differences that are important to recognize for the purpose of eliminating high
cereulide producers from food.
33
Tabl
e 11
. Cer
eulid
e pr
oduc
tion
and
affil
iatio
n of
the
B. c
ereu
s stra
ins s
tudi
ed in
this
thes
is. T
he st
rain
s wer
e gr
own
on tr
yptic
ase
soy
agar
pla
tes.
B. c
ereu
sS
perm
mic
roas
say
Che
mic
al a
ssay
Orig
in o
f the
stra
in, r
efer
ence
Stra
ins
from
food
impl
icat
edng
of c
ereu
lide
pe
r mg
of b
acte
riaor
not
impl
icat
ed w
ith il
lnes
sB
116
190
150,
190
, 230
Mea
t pas
try, c
ontro
l sam
ple
(not
food
poi
soni
ng) F
inla
nd (P
aper
II)
B20
336
025
0, 3
60, 3
80R
ice
mus
h, c
ontro
l sam
ple
(not
food
poi
soni
ng) F
inla
nd (P
aper
II)
B20
810
010
0, 1
20, 1
20C
ake,
food
poi
soni
ng, F
inla
nd (u
npub
lishe
d)F4
810/
7232
032
0, 3
80, 4
10Fo
od p
oiso
ning
, UK
(Tur
nbul
let a
l., 1
979)
B 3
4730
032
0, 3
50, 3
50P
asta
dis
h, fo
od p
oiso
ning
, Fin
land
(P
aper
III)
B30
810
0090
0, 1
000,
150
0R
isot
to, f
ood
pois
onin
g, F
inla
nd (A
petro
aie
et a
l., 2
005)
B41
250
042
0, 4
50, 5
00C
ake,
food
poi
soni
ng, F
inla
nd (A
petro
aie
et a
l., 2
005)
F588
1/94
500
320,
400
, 450
Frie
d ric
e, U
K (A
nder
sson
et a
l., 1
998)
B11
7<
0.9
< 0.
2M
eat p
astry
, con
trol s
ampl
e (n
ot fo
od p
oiso
ning
) Fin
land
(Pap
er II
)F5
28/9
4<
0.9
< 0.
2B
eef c
how
mei
n an
d ric
e, U
K (P
irttij
ärvi
et a
l., 1
999)
Env
ironm
enta
l iso
late
sLK
T I/1
400
350,
400
, 500
Filli
ng m
ater
ial o
f a b
uild
ing
with
moi
stur
e da
mag
e, F
inla
nd (A
petro
aie
et a
l., 2
005)
7pk4
5030
, 50,
80
Indo
or w
all o
f a h
ospi
tal w
ith m
oist
ure
dam
age,
Fin
land
( A
petro
aie
et a
l., 2
005)
NS5
815
0090
0, 1
000,
110
0Li
ve N
orw
ay s
pruc
e, F
inla
nd (H
alla
ksel
aet
al.
, 199
1)N
S88
1500
1000
, 150
0, 1
700
Live
Nor
way
spr
uce,
Fin
land
(Hal
laks
ela
et a
l., 1
991)
NS1
1510
0070
0, 9
00, 1
000
Live
Nor
way
spr
uce,
Fin
land
(Hal
laks
ela
et a
l., 1
991)
NS1
1710
0011
00, 1
200,
120
0Li
ve N
orw
ay s
pruc
e, F
inla
nd (H
alla
ksel
aet
al.
, 199
1)
Hum
an a
nd c
linic
al is
olat
esN
C74
0130
038
0, 4
00, 4
00Fo
od p
oiso
ning
pat
ient
, Jap
an (A
gata
et a
l., 1
994)
RIV
M B
C 0
0067
2020
, 25,
40
Hum
an fa
eces
, The
Net
herla
nds
(Ape
troai
eet
al.
, 200
5)R
IVM
BC
000
6840
60, 8
0, 8
0H
uman
faec
es, T
he N
ethe
rland
s (P
aper
I)R
IVM
BC
000
7510
010
0, 1
20, 1
50H
uman
faec
es, T
he N
ethe
rland
s (A
petro
aie
et a
l., 2
005)
IH 4
1385
105,
5, 1
0D
ialy
sis
fluid
of d
ialy
sis
patie
nt, F
inla
nd (A
nder
sson
et a
l., 1
998;
Ehl
ing-
Sch
ulz
et a
l., 2
006)
B. c
ereu
s ty
pe s
train
ATC
C 1
4579
T<
0.9
< 0.
2
34
5.3 Method for direct extraction and analysis of cereulide in foods
We designed a method for extracting and analysing the emetic toxin, cereulide, from food and
applied this method in paper II to industrially manufactured bakery products (Table 12). The
assay developed, based on solvent extraction, was optimized using a robotized extraction
instrument. The best yield (> 70%) was obtained by extracting the bread with methanol-
pentane (1:1) at a temperature of 100 °C and pressure of 103.4 bar (Paper II, Table 1).
For assessment of the toxicity of the food extracts, we used the rapid boar sperm microassay
described in Paper I. It was possible to quantitatively measure cereulide in extracts of food,
containing a myriad of substances, when cereulide-specific molecular ions were used (paper
II) to minimize matrix interference. Internal calibration standards were spiked into the food
matrix. The calibration curve was close to linear from 0.01 to 10 μg of valinomycin per ml in
the bread extract. The detection limit was 2 ng of cereulide per g of bread. I used
valinomycin as an internal and/or also as an external standard to assess the efficiency of
extraction from the complex matrices such as food. The specific ions that I used to quantify
the internal standard valinomycin were m/z 1111-1113, 1128-1130, 1133-1135 and 1149-
1151.
Our developed method, specific for cereulide and based on LC-MS, has since been applied in
various materials in many research projects at our laboratory: minced meat pasta food:
(Pirhonen et al. 2005); infant food formulas (Shaheen et al., 2006) and paper pulp (Hoornstra
et al., 2006). We also initiated collaboration with laboratories in several countries. Our
method for foods was adopted by Carlin et al. (2006); Rajkovic et al. (2006b); Svensson et al.
(2006); and Thorsen et al.( 2006).
After our work was published in 2003, Hormaz bal et al. (2004) described an LC-MS-based
quantification of cereulide in two foods: figs and rice. Their method of extraction was more
elaborate than ours. They reported as the detection limit 1 ng of cereulide per g of food and as
the quantification limit 2 ng of cereulide per g of food. This is similar to the results of our
method: 2 ng of cereulide per g of food. Their extraction method was based on a mixture of
acetone-tetrahydrofuran, methanol and water. The organic layer was separated from the
aqueous layer by adding chloroform. The method thus required solvents highly toxic to
humans (tetrahydrofuran, chloroform), whereas we used less hazardous solvents. Hormaz bal
35
et al. (2004) used only one specific ion, the NH4+ adduct, m/z 1170.9, for cereulide detection.
We used four different cereulide-specific ions (chapter 5.2), corresponding to adducts of K+,
Na+, H+ and NH4+. In our opinion this is needed for accurate assays, because the adduct ratios
may vary between different food matrices and analyses.
Recently, we began to use ethanol as the solvent for cereulide extraction. Ethanol is less toxic
to humans than methanol and less explosive than pentane. Ethanol turned out to be at least
equally as effective as the previously used solvents. In this method, the plain food (usually 1-
10 g) is flooded with ethanol and allowed to equilibrate in a stationary position in a closed jar
at room temperature (21-23 °C) overnight. The ethanol phase is then harvested and is
evaporated to dryness at 50 °C. After all liquid evaporated, the residue is dissolved in 1 ml of
ethanol or methanol.
5.4 Mining for cereulide producers from food
We examined various bakery products (not implicated with illness): meat pastry, rice pastry,
white bread and whole-grain wheat bread for the presence of toxin-producing strains of B.
cereus (Table 12). Of each food item, five parallel products were acquired from
manufacturers or from the consumer markets. Before analysis, the foods were preheated
(72 °C, 5 min) to activate the spores and then stored for 4 d at room temperature (21-23 °C).
Before and after storage, parallel aliquots were combined, mixed and streaked on bovine
blood agar and cultivated for 1 d at 28 °C. Colonies with a morphological appearance
resembling that of B. cereus (i.e. sensu lato), were selected for the study. From the two types
of pastry, 20 colonies with B. cereus-type morphology were selected for toxicity analysis
using the rapid boar sperm microassay, the LC-MS method (cereulide) and by commercial kit
for haemolysin BL. From the meat pastries 14 of the 20 tested (70%) weakly haemolytic
isolates were toxic in the sperm assay and one (5%) of the 20 from the rice pastries. LC-MS
analysis confirmed that the toxic compound was indeed cereulide. The diarrhoeal HBL
enterotoxin was produced by other B. cereus strains from the same rice and meat pastries.
These results show that industrially prepared pastries may contain cereulide producers. In
paper II we showed that the rice and meat pastries supported cereulide production when the
producer strains were present. In contrast to meat and rice pastries we found no toxic B.
36
cereus from any of the breads (Table 12). Pirhonen et al. (2005) also described a food that
contained both diarrhoeal (HBL) and emetic toxin (cereulide) producers.
The plate-culturing medium first used for determining the pathogens from foods incriminated
with food poisoning incidents is often blood agar incubated at 30 °C (Parry et al., 1983) or at
35 °C (FDA, 1998). The colonies used for pathogenicity testing are usually picked from
overnight-grown plates based on the characteristic B. cereus colony morphology, size and
zone of clear haemolysis (reviewed in the doctoral thesis of Pirttijärvi, 2000). We determined
that emetic toxin-producing isolates were found exclusively among colonies with low
(clearing zone of ≤ 2 mm) or no haemolytic activity (i.e. no clearing zone) on plates with 5
vol % of defibrinated bovine blood (Paper I, Figure 1). We found not a single isolate with a
wide, clear zone of haemolysin that would produce cereulide, although > 200 isolates were
tested (Paper I). This demonstrated that only by choosing the strongly haemolytic colonies
from the primary plate culture are the cereulide-producing strains likely to be excluded.
37
Tabl
e 12
. Min
ing
for
spor
e-fo
rmin
g to
xic
bact
eria
and
B. c
ereu
s fr
om in
dust
rially
man
ufac
ture
d ba
kery
pro
duct
s fo
r da
y of
pur
chas
e an
d af
ter
stor
age
of 4
d a
t 21
-23°C
. Mos
t spo
re-fo
rmin
g ba
cter
ia b
elon
g to
the
B. c
ereu
s s.l
.gro
up
(bas
ed o
n ty
pe o
f co
lony
mor
phol
ogy)
. Cer
eulid
e
prod
uctio
n w
as m
easu
red
with
the
boar
spe
rm m
icro
assa
y an
d th
e pr
esen
ce o
f cer
eulid
e w
ith th
e LC
-MS
anal
ysis.
Ent
erot
oxin
hae
mol
ysin
BL
prod
uctio
n w
as d
etec
ted
with
an
imm
unoa
ssay
(O
xoid
kit)
.
Tota
l col
ony
coun
t of
spor
e-fo
rmin
g ba
cter
iaC
olon
y co
unt o
fB. c
ereu
s s.
l.Is
olat
ed e
met
ic s
trai
ns*/
all
Isol
ated
dia
rrho
eal s
trai
ns**
/(c
fu/g
)(c
fu/g
)te
sted
B. c
ereu
s s.
l. st
rain
sal
l tes
ted
B. c
ereu
s s.
l. st
rain
s
Bak
ery
prod
ucts
on d
ay 0
on d
ay 4
afte
r sto
rage
on
day
0
on d
ay 4
afte
r sto
rage
4 d
4 d
Whi
te b
read
10-1
0010
310
100
0/ 7
0/ 5
Who
le g
rain
-whe
at b
read
10
-100
103
00
0 /0
0/ 0
Mea
t pas
try10
310
810
010
814
/ 20
1/ 2
0R
ice
past
ry10
310
810
010
8 1
/ 20
2/ 2
0
*) p
ositi
ve fo
r bot
h to
xici
ty (s
perm
test
) and
the
prod
uctio
n of
cer
eulid
e
**) p
ositi
ve in
the
HB
L te
st
38
5.5 Cases of emetic B. cereus food poisoning
Using the method newly developed in this thesis for analysing cereulide directly from foods
(Paper II), we analysed the remains of a pasta dish consumed by two adult persons
subsequently taken sick by emetic illness. Local authorities were able to rescue the remains of
the poisonous meal and first plated the suspected food as usual (chapter 5.4) and selected a
few B. cereus resembling colonies for preparing pure cultures. These strains all produced
diarrhoeal toxins; none produced cereulide. However, the illness symptoms of the affected
persons indicated the emetic syndrome and our laboratory was therefore called in to restudy
this food in collaboration with the Finnish National Veterinary and Food Research Institute
EELA (since 2006 renamed the Finnish Food Safety Authority, EVIRA). A total of 122 B.
cereus isolates were randomly sampled from this food and over half of these (68%) produced
the emetic toxin, as shown by the rapid boar sperm microassay (Paper I; Pirhonen et al.,
2005). The remains of the consumed meal were then solvent-extracted directly by both the
manual (Andersson et al., 1998) and the robotized (Paper II) protocol. The same type of
meal, not associated with the food poisoning, was purchased from a local store for reference.
The toxicity titers of both dishes were determined by the boar sperm microassay.
The results showed that the manually prepared extract of the illness-incriminated food
contained 1 - 2 μg of cereulide equivalents of the emetic toxin per g of the suspected food.
The extract prepared with the robotized method, contained 1.5-3 μg of cereulide equivalents
of the emetic toxin per g of the food. When the substance, cereulide, was quantitated by the
calibrated LC-MS method, 1.4 μg g-1 were found in the manually prepared extract and 1.7 μg
g-1 in the robotized extract. Mass spectrum of robotized extracts is shown in Figure 3. The
reference food contained cereulide below the detection limit of the LC-MS method, which for
that food was 0.01 μg cereulide g-1. The illness-affected persons thus had consumed ~170 µg
of cereulide per each 100 g of the ingested food. The illness-causing dose thus may have
been 8 μg kg-1(60 kg), more likely 2 - 5 μg kg-1, if the amount of the food actually ingested
did not exceed 100 - 200 g per person. Our results thus show that humans are very sensitive
to cereulide, as are the rhesus monkey and musk shrew (Review of the literature, Table 6) and
much (50-100 ×) more sensitive than mice.
39
Figure 3. Mass spectrum of cereulide in the poisonous meal. The main adduct of the
molecular ion was NH4+, with m/z values 1171.4, followed by M+K+ (1191.9) and M+H+
(1153.8). The Na+ adduct was the smallest and is not visible in the figure.
527.4
1171.4
1191.9
1. +MS, 3.8-4.8min
0.00
0.25
0.50
0.75
1.00
1.25
1.50
7x10Intens.
500 600 700 800 900 1000 1100 1200 1300 m/z
1153.8
40
5.6 The case of acute liver failure
We analysed the cereulide contents in a human liver sent to us from Belgium. The liver had
been removed from an infant patient in a Belgian hospital. At the time of arrival, the liver had
already been preserved in formaldehyde to prevent decay during transport. The infant, 11
months of age, suffered from liver failure (steatotic liver) with a suspected association with
food poisoning. The child had been in hospital care for many days before removal of the liver.
The physician suspected possible food poisoning by emetic toxin-producing B. cereus, but no
food was available for analysis.
We used porcine liver (purchased from the local store) as a model to design a method for the
extraction and analysis of cereulide from a formalinized liver. The method we developed was
as follows:
1. Washing
2 g of the fresh liver were soaked in 20 ml of distilled water at room temperature
overnight. The water was changed and the formalinized liver tissue stored
another night at room temperature.
2. The water was drained and the liver tissue dried at 60 °C.
3. The liver tissue was ground under liquid nitrogen in a mortar.
4. The dry liver powder was flooded with 10 ml of ethanol and incubated for 2 d at
room temperature.
5. The ethanol phase was harvested and evaporated to dryness at 50 °C. When all
liquid had evaporated, the residue was dissolved in 1 ml of methanol.
6. The methanol extract was stored at -20 °C until LC-MS analysis. The limit of
detection after this extraction protocol by LC-MS was 5 ng of cereulide per g
fresh wt of liver (obtained by calibration with cereulide-spiked formalinized
porcine liver).
We analysed the child´s liver using this method, but the concentration of cereulide, if present,
remained below the detection limit of 5 ng g-1. Afterwards we realized that the lipophilic
toxin, cereulide (log Kow = 6), could already have migrated from the liver through the blood
stream and into the body fat, which should have been available for analysis.
41
5.7 Analysis of the toxicity target of cereulide in mammalian somatic and
germ cells
In our research group, boar spermatozoa have been used for detecting cereulide toxicity since
1998. The question arises whether the high toxicity of cereulide to boar spermatozoa is
dependent on the animal species or whether it is specific for haploid, gametic cells like
sperms. To answer this question we investigated the cereulide sensitivities of bovine and
porcine sperm cells (Paper III). Both sperms were obtained as commercial products,
purchased from the suppliers of sperm for farm use. The responses to cereulide of these sperm
cells were compared with those of the commercially available human somatic cell lines
cervical cancer (HeLa), colon carcinoma (Caco-2), lung carcinoma (Calu-3) and a research
cell line, neural cell (Paju). HeLa is one of the most widely used cell lines in the world. The
Caco-2 cells were used to model the contact of cereulide in food with the human digestive
epithelia tract and Paju cells to assess the potential for neurotoxicity. The Calu-3 cells were
used to model exposure to inhaled toxin. This was done because cereulide-producing bacteria
are known to occur in moisture-damaged buildings where the occupants suffer from building-
related illness (Andersson et al., 2002).
The effects of cereulide on the membrane potentials of the mitochondria (Δψm) were
visualized by staining with the membrane potential-sensitive dye JC-1. The results showed
that the threshold concentration of cereulide for dissipating the Δψm was similar in the four
types of cultured human somatic cells and in the boar sperm cells: 2 ng of cereulide per ml.
The sperm cells in bull semen tolerated over 100 times more cereulide than did those in the
boar semen. Commercially available bull semen is sold in frozen form and stored in liquid N.
As such it contains an extender with cryoprotectants. In contrast, boar semen is sold unfrozen
and its extender contains no cryoprotectant. The toxicity of cereulide may have been
attenuated by the freeze-preserving additives rich in protein and lipid. Lipids are known to
attenuate the toxicity of lipophilic bioactive substances as explained by Seibert et al. (2002).
Cereulide is highly lipophilic with a log Kow = 6 (Teplova et al., 2006). The cell density,
exposure conditions, cultivation method (suspension or monolayer) affect the sensitivity of
the exposed cells to toxins. It is also known that malignant cell lines may be less sensitive
than primary cells (Paananen et al., 2002; Teplova et al., 2004 and Andersson et al., 2007).
42
My results indicate that cereulide is a rapidly acting (minutes to hours) universal poison, to
which all mammalian cells are sensitive, germ cells as well as somatic cells.
5.8 Cereulide production under different environmental conditions
This chapter deals with factors affecting cereulide accumulation in artificial media or in foods
and seeding with emetic strains of B. cereus.
5.8.1 Cereulide production by emetic B. cereus in laboratory cultivation
media (cereulide contents of the harvested bacterial biomass)
My research showed that the content of cereulide in B. cereus biomass was strongly
modulated by the growth environment of the bacterium; Table 13 summarizes the results.
The cereulide content of B. cereus biomass harvested from rich agar media (tryptic soy agar,
brain heart infusion agar and blood agar) was high, 220-450 µg g-1 of biomass wet wt. Lower
amounts of cereulide (22 - 71 µg g-1 wet wt) accumulated when the same emetic B. cereus
strains were grown on medium-rich agar (MYP, R2 agar or rice-water agar). B. cereus grew
well on MYP agar (composed of mannitol, egg yolk and polymyxin B agar), equal to levels
found in the richest media. On R2 agar (composed of yeast extract, peptone, casamino acids,
dextrose, starch, sodium pyruvate, dipotassium phosphate and magnesium sulphate), as well
as on rice-water agar, B. cereus formed less dense, but clearly visible, colonies. Adding L-
leucine and L-valine (0.3 g l-1) stimulated cereulide production 10 - 20 - fold on R2 and rice-
water agar (Paper IV). This increase in cereulide production was induced by the free amino
acids (L-leucine and L-valine) but not peptide-bonded amino acids. This was documented by
adding peptone containing similar amounts of peptide-bonded L-leucine and L-valine: there
was no effect on cereulide production. These amino acids, L-leucine and L-valine, are also
approved food supplements (flavour modifiers) in the USA [http://jecfa.ilsi.org
/evaluation.cfm (4.10.2007)] and in the EU (European Commission, 2006). The Scientific
Panel has not considered food supplement effects for microbial toxin production (European
Commission, 2006).
43
5.8.2 Time course of cereulide production
The question often asked is when cereulide production begins and finishes in different media.
Figure 4 shows the time course of cereulide accumulation by independent isolates of emetic
B. cereus from live Norway spruce (Picea abies). The strains were isolated with aseptic
equipment from live trees in the forest during the coldest period in winter (Hallaksela et al.,
1991). B. cereus strains NS 85, NS 88, NS115 and NS 117 were grown on tryptic soy agar 7 d
at room temperature (21-23 °C) and the biomass obtained analysed for cereulide. In two of the
strains, the concentrations of cereulide in the biomass continued to increase for 3 d and in two
other strains for 6 d, indicating that individual strains, although of the same origin, may have
different kinetics of cereulide production. Decrease in the cereulide content of the B. cereus
strains when the cultures became 6 - 7 d old indicates that cereulide may be autodegraded by
its producer strains.
Table 13 also shows the results for liquid laboratory media and milk. Cereulide production by
B. cereus strains B116, B203 and F4810/72 in trypticase soy broth mainly started 16 h after
inoculation at 22 °C. After 65 h the concentration of cereulide in the broth rose to 3-6 µg
ml-1. Sporulation of B. cereus on tryptic soy agar begins after ~ 48 h. These results are in line
with those of Häggblom et al. (2002), who showed that cereulide accumulation in broth
cultures started as soon as the culture reached the stationary phase, i.e. before the culture
sporulated, and then remained at a plateau concentration for the subsequent 24 h.
Based on our results the final concentration of cereulide was probably reached overnight on
rich solid media, such as brain heart infusion agar, whereas in tryptic soy broth cereulide
production only started after 16 h of incubation at 22 °C. The results further indicate that the
onset of cereulide production occurred sooner on solid culture media than in liquid medium -
explaining why cereulide food poisonings apparently have never been reported for liquid
foods.
44
5.8.3 Cereulide production in foods
I investigated foods to determine their amenability to accumulate cereulide (Table 13). I
found that rice pastry and meat pastry (seeded with B. cereus strains F4810/72, B116 or
B203) accumulated 0.7-5.5 µg of cereulide per g within 4 d (Table 13). The pastries contained
rice and proteinaceous additives. Plain boiled rice also accumulated large amounts, 2 - 4 µg
per g of food, of cereulide. Rice alone aparently contains sufficient amounts of the essential
amino acids (threonine, leucine and valine) to maintain growth of B. cereus (naturally
auxotrophic for these amino acids) and cereulide production. I found that the cereulide-
producing strains studied (B116, B203 and F4810) survived the heating applied during
baking of pastries 20 min at 250 ºC (dry heat) and cooking of food (boiling for ~ 30 min).
Wijnands et al. (2006) found that rice- and pasta-containing dishes (ready-to-eat foods)
mostly contained ≥ 105 cfu of B. cereus per g sampled under normal retail conditions. The
dose of B. cereus inoculated in the foods in our studies, 106 cfu g-1, was therefore realistic.
I cooperated with Andreja Rajkovic by analysing the food samples from his study for
cereulide by the LC-MS method. The samples were pasta, potato puree, milk and rice
inoculated with B. cereus strains NS117 and 5964a (Rajkovic et al. 2006b). I found 2 µg g-1of
cereulide in the rice, which is similar to what I found earlier in rice (Paper IV). I found high
amounts of cereulide in the potato puree and pasta (after 48h shelving time at 28 °C) sent to
me by A. Rajkovic, 4 and 3 µg g-1, respectively, clearly showing, that these foods are
sensitive to cereulide production.
Cereulide production in milk is an interesting topic. I found that no cereulide (< 0.5 ng ml-1)
was produced in shaken consumer skim milk at 22 °C (Table 13), even though it had been
seeded with 106 cfu of cereulide producer strains 4 d earlier. My studies with Andreja
Rajkovic (2006b) showed that no cereulide accumulated in shaken milk at 28 °C after 48 h.
However, I found in the same study that the same strains of B. cereus, NS117 and 5964a,
produced 1 µg ml -1 in whole consumer milk that had been shelved stationary.
Agata et al. (2002) reported different results from Japan: shaken milk was more toxic in the
HEp-2 cell assay (0.64 µg cereulide equivalents per g ) than stationary (< 0.01 µg g -1)
incubated milk. In the protocol of Agata et al. (2002), the B. cereus was grown in milk for
45
24 h, centrifuged, the supernatant collected and autoclaved and the toxicity titer measured in
the supernatant. Their results (Table 4, Agata et al., 2002) show that after 24 h the B. cereus
strains in the stationary incubated milk culture was still growing, whereas the shaken culture
had already attained the maximal cfu content. In my study B. cereus was grown for 4 d (Table
13) or 48 h (study done with A. Rajkovic, 2006b ). At these times, the static cultures should
also have been fully grown. A further point needing emphasis, is that cereulide is insoluble
in water (log Kow = 6, Table 6). Consequently, the toxin produced will likely remain bound to
the bacterial cells or their debris, or (in the case of milk) accumulate in the fat and float in the
supernatant. Therefore the outcome for skim milk will be different from that of whole milk. In
my study, I applied no centrifugation step and the whole-milk sample was extracted with an
organic solvent suitable for solubilizing cereulide.
The first report on B. cereus emetic toxin in milk (or liquid growth media) is that of Szabo et
al. (1991). Later, other researchers (Sakurai et al., 1994; Agata et al. 1996,1999, 2002; Finlay
et al., 1999) followed the Szabo protocol. In this protocol the liquid culture was centrifuged,
the pellet discarded and the supernatant boiled or autoclaved (to destroy the heat-labile toxins
and viable bacteria) before the toxicity assay. In such a protocol, the toxin bound to the
bacterial cell pellet, or to the food debris, is lost from the toxicity result. This is especially
true for skim milk, where there is no fat to retain the toxin in the supernatant. In some
protocols (Finlay et al., 1999), the centrifugation was done for 40 min at 4 °C, likely
immobilizing most of the cereulide in the hydrophobic phase and on the walls of the
centrifuge tubes. The protocol of Mikami et al. (1994) was different: the autoclaving step
preceded the centrifugation. Mikami et al., call their protocol an “improved method”, leading
to higher yields of cereulide, possibly because autoclaving lysed the cells, reducing the size
of the cell pellet and thus increased the toxin yield in the supernatant.
Our study is the first in which the substance cereulide as well as its toxic effects were
measured directly in food. All other studies published to date outside our laboratory used
indirect methods, including bioassays on toxicity of heated B. cereus or of food extracts.
Since no heat-stable toxin other than cereulide has so far been found in B. cereus from foods,
all the studies published by Agata et al. (1999, 2002) and Szabo et al. (1991) most likely
measured toxicity caused by cereulide.
46
Hormaz bal et al. (2004) published an application of our earlier described LC-MS method, in
which they determined the cereulide contents in figs and rice. These authors did not measure
the toxicity from these samples. We believe that it is important to combine the LC-MS
method with the bioassay method to be sure of the source of the toxicity. This was done for
determination of cereulide from infant food formulas (Shaheen et al., 2006).
5.8.4 Cereulide production under different atmospheres
Table 13 also summarizes the results in which cereulide productions in the same foods or
media by the same strains of B. cereus were measured under different atmospheres. These
results show that in > 99.5% N2, no cereulide was produced in liquid laboratory media or in
the two solid foods studied (rice, beans). However, the cereulide contents found after aerobic
or anaerobic incubations of B. cereus strains on tryptic soy agar plates were also highly
independent of the atmosphere: anaerobic (CO2 9-13%, O2 < 1%, in N2) or ambient air.
Finlay et al. (2002) reported that the density of B. cereus strains F4810/72, F3748/75,
F3744/75, F4562/75, F4552/75 and F2427/76, measured as viable counts after 24 h at 30 °C
in skim milk medium, remained lower (P < 0.01) under anaerobic conditions than under
aerobic conditions. Toxicity, presumably due to cereulide, was nondetectable in the anaerobic
cultures, even though the viable counts were consistently > 106 cfu ml-1 which should have
been sufficient for producing measurable amounts of cereulide (Finlay et al., 2002).
Rajkovic et al. (2006a) found that when the atmosphere contained less than 1.6 vol % O2 in
N2, no cereulide was produced by B. cereus growing on solid medium (tryptic soy agar 24 h,
28 °C), but ample cereulide was produced, about 1 µg g-1 biomass wet wt, on the same plates
in an atmosphere of > 4.5% O2 in N2. The B. cereus strains used in that study were NS117
(Finnish spruce tree isolate from our laboratory) and 5964a (food isolate from a fatal case of
B. cereus poisoning in Belgium).
Based on our results and those published by other researchers, the role of O2 in cereulide
production by B. cereus is not a simple one. An N2 atmosphere in the absence of CO2 did not
allow cereulide production in the absence of O2 but did in the presence of CO2. However, the
CO2 or lowering of the redox -potential of the growth environment could have promoted
47
toxin production in facultative anaerobic B. cereus. Further studies are necessary along this
line.
48
Tabl
e 13
. Acc
umul
atio
n of
cer
eulid
e in
B. c
ereu
s bi
omas
s whe
n gr
own
in d
iffer
ent c
ultiv
atio
n m
edia
and
in fo
ods t
hat w
ere
seed
ed w
ith e
met
icB.
cer
eus s
train
s. B
116
was
isol
ated
from
mea
t pas
try (F
inla
nd, a
con
trol s
ampl
e), B
203
from
rice
por
ridge
(Fin
land
, a c
ontro
l sam
ple)
F48
10/7
2fro
m a
food
poi
soni
ng c
ase
(UK
). A
ll re
sults
are
giv
en a
s mea
ns o
f tw
o or
thre
e in
depe
nden
t rep
licat
e cu
lture
s of t
he sa
me
strai
n.In
cuba
tion
at 2
2°C
B. c
ereu
sst
rain
Gro
wth
sub
stra
teda
ys o
f gro
wth
atm
osph
ere
B11
6B
203
F481
0/72
Bio
mas
s of
B. c
ereu
sha
rves
ted
from
Cer
eulid
e m
easu
red
in th
e ba
cter
ial b
iom
ass
(µg/
g )
Tryp
tic s
oy a
gar
4am
bien
t28
045
035
0Tr
yptic
soy
aga
r4
CO2 9
-13%
, O2
< 1
% in
N2
310
400
320
Brai
n he
art i
nfus
ion
agar
4am
bien
t23
536
045
0Bl
ood
agar
4am
bien
t22
028
030
0R
2 ag
ar4
ambi
ent
4074
71M
anni
tol e
gg y
olk
poly
myx
in B
aga
r4
ambi
ent
2550
22R
ice-
wat
er a
gar
4am
bien
tN
DN
D28
Liqu
id m
edia
(sh
akin
g 12
0 rp
m)
Cer
eulid
e m
easu
red
in th
e liq
uid
(µg/
ml)
Con
sum
er s
kim
milk
(0%
fat)
4am
bien
t0.
020
0.05
Con
sum
er s
kim
milk
(0%
fat)
4>
99.5
vol
% N
20
00.
08Tr
yptic
ase
soy
brot
h4
ambi
ent
5 3
.5 5
.5Tr
yptic
ase
soy
brot
h4
> 99
.5 v
ol %
N2
0.01
00
B. c
ereu
s-in
ocul
ated
food
sC
ereu
lide
in th
e fo
od (µ
g/g)
Whi
te b
read
4am
bien
t0.
02 0
.03
0.03
Who
le-g
rain
whe
at b
read
8am
bien
t0.
010.
020
Rye
bre
ad21
ambi
ent
00.
01 0
.01
Mea
t pas
try, d
ough
4am
bien
t 0
.80.
60.
3M
eat p
astry
, filli
ng4
ambi
ent
0.7
4.2
5.5
Ric
e pa
stry
4am
bien
t 1
.8 1
.5 1
.55
Bean
s, b
oile
d4
ambi
ent
ND
1.6
0.7
Bean
, boi
led
4>
99.5
vol
% N
2N
D0.
10.
06R
ice
, boi
led
4am
bien
t3
24
Ric
e, b
oile
d4
> 99
.5 v
ol %
N2
0.01
0.02
0.01
ND
= no
t det
erm
ined
49
Figu
re 4
.Pro
duct
ion
of c
ereu
lide
byB.
cer
eus s
train
s (iso
late
d fro
m N
orw
ay sp
ruce
) NS5
8, N
S88,
NS1
15 a
nd N
S117
. The
stra
ins w
ere
grow
n
aero
bica
lly o
n try
ptic
soy
agar
at 2
2°C
. The
cer
eulid
e co
nten
ts o
f ind
epen
dent
repl
icat
es (t
hree
) wer
e m
easu
red
with
LC
-MS.
Cer
eulid
e pr
oduc
tion
ofB
. cer
eus
str
ains
0
500
1000
1500
2000
2500
3000
12
36
7
days
of c
ultiv
atio
n
cereulide contents ofB. cereus harvested(ng/mg biomass)
NS5
8N
S88
NS1
15N
S117
50
6. Conclusions
When I initiated my doctoral research in 2001, most published studies on B. cereus were
focused on the production of various diarrhoeal enterotoxins. Here I focused on production of
the B. cereus emetic toxin, cereulide, a known mitochondriotoxin. Significant outcomes of my
work include the following:
1. We developed a rapid in vitro method to screen for the presence of the heat- stable B.
cereus toxin for a large numbers of strains in a short time. This method is based on the rapid
(5 min) effect of cereulide on boar sperm cells. We found that the toxin-positive B. cereus
strains always had a phenotype of poor haemolysis on blood agar. This revelation was used
to preselect the poorly haemolytic colonies for toxin analysis. This is contrary to the current
practice in most laboratories, where the haemolytic B. cereus colonies are preferred.
2. We studied the toxicity threshold of cereulide for the human HeLa, Caco-2, Paju and Calu-
3 cell lines. We found that the toxicity endpoint of cereulide for boar sperm cells and human
cells was similarly low, showing a detection limit of 2 ng of cereulide per ml of cells. This
indicates that the boar sperm assay is suitable for in vitro assessment of possible effects on
human cells by extracts suspected of containing the mitochondrial toxin, cereulide, of B.
cereus.
3. We designed a method for quantitative extraction of the B. cereus emetic toxin not only
from the biomass of laboratory-grown B. cereus, but also directly from foods. Cereulide is a
highly lipophilic substance and is practically insoluble in water. The novel extraction protocol
is based on organic solvents. The extraction was optimized (100 °C, 103.4 bar) to effectively
solubilize cereulide from bacterial biomass and from food.
4. The extracted cereulide was separated from other constituents by LC and then quantified
based on the m/z values of cereulide-specific NH4+, H+, Na+ and K+adducts. The bioassay for
toxicity was performed on the same extract, using the boar sperm microassay. This double
protocol verified that cereulide is the toxin identified and that it preserved its biological
activity (toxicity) despite the aggressive extraction method.
51
5. Using the new method for cereulide quantification, we were able to disclose the dose of
cereulide causing illness for healthy adult persons. We analysed the actual remains of the
meal implicated in an outbreak of cereulide poisoning, using the new boar sperm microassay
and the novel chemical assay based on LC-MS. Both methods showed that pasta contained
1.5 - 1.7 μg of cereulide per g. Ingestion of 100 g of such food means exposure to 150-170 μg
of cereulide. My report was the first worldwide in which the dose causing a serious acute
vomiting syndrome in humans was established. My results showed that the acute illness-
causing dose is lower for cereulide than that for any other known microbial heat-stable toxin.
6. I studied several industrially manufactured foods to determine their susceptibility to
accumulated cereulide. I found that rice, rice-containing pastries and beans accumulated high
concentrations of cereulide, 0.3 - 5.5 µg g-1, when stored at nonrefrigerated temperatures for
up to 4 d. My results show that if emetic B. cereus strains are present in food, the risk of food
poisoning cannot be overlooked when nonrefrigerated products, such as bakery products, are
eaten days after manufacture.
7. A direct cereulide-specific assay made it possible to identify environmental factors
promoting or preventing the production of this toxin. I found that B. cereus emetic strain
(F4810/72) produced 450 µg of cereulide per g of cells (wet wt) on brain heart infusion agar
during 4 d at room temperature. Under similar conditions, B. cereus (F4810/72) produced
only 22 µg of cereulide per g of cells (wet wt) on mannitol egg yolk agar, 28 µg on rice-water
agar and 71 µg on R2-agar. Adding the free amino acids L-leucine and L-valine stimulated
cereulide production on oligotrophic R2 agar and/or rice-water agar 10 - 20 fold.
Interestingly, adding meat peptone (5 g l-1), containing the same amount of (peptide-bonded)
amino acids (0.3 g L-1) in the same medium promoted growth of the toxin producer, but had
no significant effect on cereulide production. L- valine and L-leucine are approved food
supplements and widely used as free amino acids in food technology.
8. Storage of B. cereus cultures or foods under N2 atmosphere (> 99.5 vol % of N2) prevented
the production of cereulide for 4 d. But when CO2 was present, the absence of O2 did not
prevent the production of cereulide. This may indicate that CO2 or lowering of the redox
potential promoted toxin production, but further studies are needed.
52
9. The actual production of cereulide was strongly strain-dependent; 5-1700 ng of cereulide
per mg of B. cereus biomass (wet wt). Therefore, it is not possible to predict the toxic
potential of any foods based only on the presence and density of the cereulide synthase gene
as measured by quantitative PCR.
10. Emetic toxin-producing B. cereus strains can readily be detected in rice-containg pastries
several days after baking. I found that several cereulide-producing strains (B116, B203 and
F4810) survived the heating applied during baking of pastries 20 min at 250 ºC and cooking
of food (boiling for ~ 30 min).
53
7. Tiivistelmä
Suomessa rekisteröityjen ruokamyrkytysepidemioiden määrä on vaihdellut samoissa luvuissa
rekistereiden koko pitoajan, 40-90 epidemiaa ja 1000-9000 ruoasta tai juomavedestä
sairastunutta henkilöä vuosittain. Näin siitä huolimatta että hygienian keinot, mm. kylmäketju
on tuona aikana parantunut. Vuoteen 2004 saakka salmonella ja sitten kampylobakteeri olivat
bakteeriepidemioiden pääasialliset aiheuttajat, mutta viime vuosina 2005–2006 Bacillus
cereus nousi yleisimmäksi. Samantapainen kehitys alkoi mm. Saksassa jo 1990 luvulla. Yksi
syy tähän kehitykseen saattaa olla Bacillus cereuksen tuottaman oksetustautia aiheuttava
toksiini, kereulidi. Bacillus cereus on luonnossa ja elintarvikkeiden raaka-aineissa hyvin
yleinen bakteeri. Toisin kuin salmonellat ja kampylobakteerit, se tuottaa itiöitä jotka kestävät
pastöroinnin ja keittämisen sekä toksiinia joka kestää jopa höyryautoklavoinnin. Bacillus
cereus itiöt aiheuttavat ruokamyrkytysriskin kuumennetuissa elintarvikkeissa ja ruoissa joita
ei syödä valmistuspäivänä, koska ruoan jäähtyessä itiöt muuttuvat kasvullisiksi
bakteerisoluiksi ja voivat tuottaa toksiineja. Tämän väitöskirjatyön aihe oli kereulidi ja sitä
tuottavien Bacillus cereus kantojen tunnistaminen, mittaaminen ja kereulidin tuottoon
vaikuttavat tekijät.
Kehitin menetelmiä kereulidin mittaamiseksi suoraan elintarvikkeesta. Määrittämisen
edellytys oli toksiinin kemiallisten ja fysikaalisten ominaisuuksien tuntemus, jotta saatoin
suunnitella menetelmän toksiinin tehokkaaseen eristämiseen elintarvikkeesta ja raaka-
aineesta. Koska kereulidi ei liukene lainkaan veteen, käytin uuttokemikaalina orgaanisia
liuottimia, metanolia, etanolia ja pentaania. Leipomo- ja konditoriatuotteista uutin kereulidin
korkeassa lämpötilassa (100°C) ja paineessa (103.4 Bar). Vaihtoehtoisesti uutto voidaan
suorittaa kuivattamalla elintarvike ja uuttamalla sitä elintarvikkeen tilavuuteen nähden
kaksinkertaisessa pitoisuudessa etanolia noin 12 tuntia. Tätä menetelmää käytin mm. pastalle
ja perunasoseelle. Nestemäiset elintarvikkeet, kuten maito, voidaan uuttaa pentaaniin tai
kuivattaa ja suorittaa etanoliuutto. Nämä uuttomenetelmät ovat tärkeä parannus kereulidin
aiheuttaman ruokamyrkytysriskin tutkimukselle, sillä ennen kereulidi uutettiin niin
tuottajabakteerista kuin elintarvikkeestakin veteen jolloin kereulidi saanto oli huono ja
vaihteleva riippuen elintarvikkeen rasvaisuudesta.
54
Kun mikrobin aiheuttamaksi epäiltyä ruokamyrkytystä selvitetään, pitää osata todeta kaksi
asiaa. Ensimmäiseksi epäillyn elintarvikkeen todellinen myrkyllisyys. Monet mikrobimyrkyt,
vaikkakaan ei kereulidi, inaktivoituvat elintarvikkeen käsittelyprosessin aikana esimerkiksi
kuumentaessa tai hapottamalla etikalla. Toiseksi myrkyn kemiallinen tunnistaminen. Siis
onko kyseessä kereulidi vai jokin muu lämpökestoinen aine, esim. homemyrkky eli
mykotoksiini. Tämä tieto tarvitaan myrkyn alkuperän tehokkaaseen selvittämiseen.
Myrkyllisyyden toteamiseen kehitin työtoverini Maria Anderssonin kanssa pikamenetelmän,
jonka avulla kereulidi voidaan todeta 5-15 minuutissa. Kehittämällämme testillä voidaan
nopeasti todeta mikä mahdollisista monista elintarvikkeista oli myrkyllisyyden aiheuttaja ja
siten ehkäistä lisäsairastumisia. Myrkyn tunnistaminen kereulidiksi tapahtuu
massaspektrometrisesti. Osoitin että kun tämä tehdään käyttäen kereulidin molekyylijonien
massalukuja: m/z (± 0.3) 1153.8 (M+H+), 1171.0 (M+NH4+), 1176.0 (M+Na+) ja 1191.7
(M+K+), tunnistus on aukoton. Mikäli tuotetta ei säilytetä kylmässä ja myrkkyä tuottava
bakteeri on läsnä niin mm. retkieväinä käytetyissä liha- ja karjalanpiirakoissa muodostuu
yleisen myyntiajan puitteissa sairastumisen aiheuttavia määriä, 0.3–5.5 µg kereulidia
grammassa elintarviketta.
Koska Bacillus cereuksen esiintyminen on niin yleistä, ettei siitä ole mahdollista päästä täysin
eroon, on tärkeää tietää mitkä olosuhteet käynnistävät toksiinin tuoton. Tutkimuksissani
selvisi että kereulidin tuotto voi vaihdella 10…1000 kertaisesti, olosuhteista riippuen. Kun
elintarvike oli suljettuna astiaan, jonka kaasutila sisälsi vain typpikaasua (99.5 %), kereulidia
ei muodostunut. Sen sijaan jos läsnä oli myös hiilidioksidia, kereulidia muodostui, vaikka
happea oli vain alle 1 %. Myös lisä-aineilla oli vaikutusta kereulidin tuottoon, ainakin
laboratorio-olosuhteissa. Leusiini ja valiini moninkertaistivat kereulidin tuoton.
Peptidimuodossa nämä aminohapot ovat kaikkien proteiinien luontainen ainesosa. Yllättävää
oli että vapaassa muodossa kasvatusalustaan lisätty leusiini ja valiini moninkertaistivat
kereulidin tuoton, mutta proteiiniin sitoutuneilla aminohapoilla ei vastaavaa vaikutusta
havaittu. Sekä leusiini että valiini ovat yleisesti käytettyjä valmisruokien aromivahventeita.
Tutkimustulokseni osoittavat että nämä lisäaineet voivat aiheuttaa ruokamyrkytysriskin
vaikkeivat itse ole lainkaan myrkyllisiä.
55
8. Acknowledgements
This study was carried out in the Department of Applied Chemistry and Microbiology,
division Microbiology, University of Helsinki, during the years 2001-2007.
My work was supported by the Graduate School for Applied Bioscience- Bioengineering,
Food & Nutrition, Environments (ABS), the National Technology Agency (TEKES) project
40132/01 in 2001, the EU-project QLK1-CT-2001-00930 "BiosafePaper", and the Academy
of Finland Center of Scientific Excellence grants 2002-2007 "Microbial Resources" 2002-
2007 and the present "PhotoBioMics" grant 2008-2013.
I am grateful to the ABS school for the financial support as well as for the scientific
education. I want to thank the present ant the former ABS graduate school co-ordinators,
Laila Partanen, Suvi Ryynänen, Sanna-Maija Miettinen and Merja Kärkkäinen, for solving
many practical problems.
Many people have contributed to this thesis in different ways and I wish to express my sincere
gratitude to them all.
I have had the great priviledge to work under the firm supervision of a distinguished world-
class scientist, Professor Mirja Salkinoja-Salonen. She provided me with expert guidance and
had always time to discuss my work. Her perceptions and suggestions were invaluable during
this project. She offered me a great opportunity to work in such a high level microbiology
laboratory. Furthermore, she always looked after our laboratory safety; it was excellent. This
is very important to me because of my pregnancies during the work.
I express my sincere gratitude to Dr. Christophe Nguyen-The and Professor Willem de Vos
for carefully reviewing this thesis and giving constructive and valuable comments which
greatly improved the final, present form in my thesis.
My gratitude goes to Professor Max Häggblom who guided me into the intriguing world of
LC-MS. I also thank Max for his new ideas and expertise while the preparing manuscripts.
56
I am grateful to Professor Mikael Skurnik and Professor Tapani Alatossava, school follow-up
group members for their support.
I thank Dr. Joel Smith for donating removed liver sample and Dr. Andreja Rajkovic for
collaborating and sharing data.
I owe special thanks to my co-authors, Maria Andersson, Ranad Shaheen, Tuula Pirhonen,
Luc Wijnands, Max Häggblom, Liisa Vanne, Vera Teplova, Magnus Andersson, Päivi
Tammela, Nils-Erik Saris, Pia Vuorela and Leif Andersson for sharing their data and
expertise while preparing the manuscripts.
I thank Tuula Pirhonen for organizing the helpful Bacillus group meeatings. Vera Teplova,
Leif Andersson, Päivi Tammela and Pia Vuorela for expert advice in the use of human cell
lines and Maria Andersson for introducing me the boar sperm toxicity assay.
I express my sincere gratitude to all the present colleagues in the MSS project (Maria
Andersson, Raimo Mikkola, Jaakko Pakarinen, Terhi Ali-Vehmas, Jaakko Ekman, Douwe
Hoornstra, Camelia Constantin, Minna Peltola, Mari Raulio and Elina Rintala) and all former
colleagues for their help and creating a pleasant atmosphere to work in. I would especially
like to mention Maria Andersson because of her friendship and helping me in so many ways. I
thank to Päivi Uutela for her guidance in the LC-MS analysis and friendship. My warmest
thanks for the friendships belongs to Hanna Sinkko as well.
Jaakko Pakarinen, Elina Rintala and Vilma Rouvinen kindly assisted me at laboratory work.
James Thompson revised the language with expertise. I want to thank them all.
I am grateful to The Division of Microbiology technical staff for their kind assistance in the
laboratory and Leena Steininger, Hannele Tukiainen and Tuula Suortti for the secretarial help.
I thank Viikki Science Library for the excellent information service and the Faculty Istrument
Centre for technical services.
I thank people in the "TRA" project: Laura Raaska, Liisa Vanne and Kaarina Aarnisalo for
helpful monthly discussons when I started my work with Bacillus cereus. I also want to thank
57
the other people at VTT as well as the participating industrial companies and the EU-project
coordinator Assi Weber abd the project members for collaboration.
I also want to thank my fellow-students, especially Tiina Thure, my pair in many laboratory
courses, for willing and able to help me and being my friend.
I owe my warmest gratitude for my family for their love and support. I sincerely thank my
father Heikki for always believing and support me. He inspired me to the world of science
and shared his great knowledge and skills when ever needed. His wife, Elina, has always
supported and encouraged me in numerous ways. My sisters Kristiina and Susanna have been
a great example of how to unite academic work and family life. Susanna gave me chance to
work with a highly intresting laboratory project which was my first laboratory work place. I
am very grateful to my brother Heikki and his family for closely sharing the life of my family
and all the joy you bring us. I also thank my father-in-law Jaakko for his phone calls, parcels
and visits.
I reserve my deepest gratitude for my beloved husband Kaius for standing by me and for all
the help and love you are giving me. Luka and Lari, our wonderful little boys, without you
nothing would matter, are the sunshines of my life. Our third, still unborn child, gave me a
very good reason to finish this work.
Helsinki, January 2008
Elina Jääskeläinen
Kiitos kaikille jotka ovat myötävaikuttaneet tämän väitöskirjan valmistumiseen!
58
9. References
Abriouel H. B., Omar N. L., Lopez R. M., Canamero M. Ortega E. Galvez A. 2007.
Differentiation and characterization by molecular techniques of Bacillus cereus
group isolates from poto poto and degue, two traditional cereal-based fermented
foods of Burkina Faso and Republic of Congo. Journal of Food Protection 70:
1165-1173
Agata N., Mori M., Ohta M., Sathorn S., Ohtani I., and Isobe M. 1994. A novel
dodecadepsipeptide, cereulide, isolated from Bacillus cereus causes vacuole
formation In Hep-2 Cells. FEMS Microbiology Letters 121:31-34
Agata N., Ohta M., Mori M. and Isobe M. 1995. A novel dodecadepsipeptide, cereulide,
is an emetic toxin of Bacillus cereus. FEMS Microbiology Letters 129:17-20
Agata N., Ohta M., and Mori M. 1996. Production of an emetic toxin, cereulide, is
associated with a specific class of Bacillus cereus. Current Microbiology 33:
67-69
Agata N., M. Ohta, M. Mori, and K. Shibayama. 1999. Growth conditions of an emetic toxin
production by Bacillus cereus in a defined medium with amino acids.
Microbiology Immunology 43:15-18
Agata N., Ohta M., and Yokoyama K. 2002. Production of Bacillus cereus emetic toxin
(cereulide) in various foods. International Journal of Food Microbiology 73: 23-
27
Andersson M. A., Mikkola R., Helin J., Andersson M. C., and Salkinoja-Salonen M. 1998.
A novel sensitive bioassay for detection of Bacillus cereus emetic toxin and
related depsipeptide ionophores. Applied and Environmental Microbiology
64: 1338-1343
59
Andersson M., Mikkola R., Apetroaie C. Hoornstra D., Nieminen T. and Salkinoja-
Salonen M. 2002. Fungicidic and mitochonriotoxic Bacilli frequent in water
damaged buildings. In Proceedings of 9 th International Conference on Indoor
Air Quality and Climate 2002 Vol 1: 34-39
Andersson M. A. Hakulinen P., Honkalampi-Hämäläinen U., Hoornstra D., Lhuquenot J.
C., Mäki-Paakkanen J., Savolainen M., Severin I., Stammati A.-L., Turco L.,
Weber A.,von Wright A., Zucco F. and Salkinoja-Salonen M. 2007.
Toxicological profile of cereulide, the Bacillus cereus emetic toxin, in
functional assays with human, animal and bacterial cells. Toxicolon 49: 351-
367
Altayar M. and Sutherland A. 2005. Bacillus cereus is common in the environment but
emetic toxin producing isolates are rare. Journal of Applied Microbiology 100:
7-14
AOAC, 1995, Bacillus cereus in foods. Enumeration and confirmation. Microbiological
methods, AOAC official method 980:31, In Cunniff P. (Ed.), Official methods
of analysis of AOAC International, 16th ed. Vol 1, AOAC International,
Arlington VA, USA pp. 52-54
Apetroaie C. , Andersson M.A, Spröer C., Tsitko I., Shaheen R., Jääskeläinen E.L,
Wijnands L.M, Heikkilä R. and Salkinoja-Salonen M. S. 2005. Cereulide
producing strains of Bacillus cereus show diversity. Archives of Microbiology
184: 141-151
Beattie S. H. and Williams A. G. 2000. Detection of toxins. In Encyclopedia of Food
Microbiology (vol 1). Edited by Robinson R. K., Batt C. A. and Patel P.
D.Academic Press, San Diego, USA. pp: 141-149
60
Beecher H. and Wong A.C.L. 1994. Identification and analysis of the antigens detected by
two commercial Bacillus cereus diarrheal enterotoxin immunoassay kits.
Applied and Environmental Microbiology 60: 4614-4616
Carlin F., Fricker M., Pielaat A., Heisterkamp S., Shaheen R., Salkinoja-Salonen M.,
Svensson B., Nguyen-the C. and Ehling-Schulz M. 2006. Emetic toxin-
producing strains of Bacillus cereus show distinct characteristics within the
Bacillus cereus group. International Journal of Food Microbiology 109: 132-
138
Cronin U. P. and Wilkinson M. G. 2007. The use of flow cytometry to study the
germination of Bacillus cereus endospores. Cytometry part A 71A: 143-153
Dierick R., van Coillie E., Swiecicka I., Meyfroidt H., Devlieger A., Meulemans A.,
Hoedemaekers G., Fourie L., Heyndrickx M. and Mahillon J. 2005. Fatal family
outbreak of Bacillus cereus-associated food poisoning. Journal of Clinical
Microbiology 43: 4277-4279
Ehling-Schulz M., Fricker M., Scherer S. 2004. Identification of emetic toxin producing
Bacillus cereus strains by a novel molecular assay. FEMS Microbiology Letters
232: 189- 195
Ehling-Schulz M., Vukov N., Schulz A., Shaheen R., Andersson M. Martlbauer E. Scherer S.
2005a. Identification and partial characterization of the nonribosomal peptide
synthetase gene responsible for cereulide production in emetic Bacillus cereus.
Applied & Environmental Microbiology 71:105-113
Ehling-Schulz M., Svensson B., Guinebretiere MH., Lindback T., Andersson M., Schulz A.,
Fricker M., Christiansson A., Granum PE., Martlbauer E., Nguyen-The C.,
61
Salkinoja-Salonen M. and Scherer S. 2005b. Emetic toxin formation of Bacillus
cereus is restricted to a single evolutionary lineage of closely related strains.
Microbiology 151:183-197
Ehling-Schulz M., Guinebretiere M-H., Monthan A., Berge O., Fricker M. and Svensson B.
2006. Toxin gene profiling of enterotoxic and emetic Bacillus cereus. FEMS
Microbiology Letters 260: 232-240
European Commission, 1993. Council Directive 93/88/EEC modifying Directive 90/697/EEc
on the protection of workers from risks related to exposure to biological agents
to work. Official Journal of European Communion. L268 of 29.10.1993, pp:
71-82
European Commisson, 2006. Opinion of the Scientific Panel on Food Additives, Flavourings,
Processing Acids and Materials in contact with Food (AFC) on request from the
Commussion related to: Flavouring Group Evaluation 26: Amino acids from
chemical group 34. Commission Regulation (EC) No 1565/2000 of July 2000.
Adopted on 4.5.2006. The EFSA Journal 373: 1-48
Fagerlund A., Ween A., Lund T., Hardy S.P. and Granum P.E. 2004. Genetic and functional
analysis of the cytK family of genes in Bacillus cereus. Microbiology 150: 2689-
2697
Fagerlund A. Brillard J. Furst R. Guinebretiere MH. Granum PE. 2007. Toxin production in a
rare and genetically remote cluster of strains of the Bacillus cereus group. BMC
Microbiology 7:43
Finlay W. J., Logan N.A., and Sutherland A.D. 1999. Semiautomated metabolic staining
assay for Bacillus cereus emetic toxin. Applied and Environmental
Microbiology 65:1811-1812
62
Finlay W.J., Logan N.A., Sutherland A.D. 2000. Bacillus cereus produces most emetic toxin
at lower temperatures. Letters in Applied Microbiology 31: 385-389
Finlay W.J.J., Logan N.A. and Sutherland A.D. 2002. Bacillus cereus emetic toxin production
in relation to dissolved oxygen tension and sporulation. Food Microbiology 19:
423-430
Food and Drug Administration (FDA), 1998. Bacillus cereus. In: Bacteriological Analytical
Manual. 8th Edition, revision A, AOAC International, Gaithersburg, MD, USA,
pp: 14.01-14.08
Forsythe S. J. 2000. Basic aspects. In: The microbiology of safe food. Blackwell Science.
Edited by Forsythe S. MA, USA, pp: 10-52
Fricker M., Messelhäuβer, Busch U., Scherer S., Ehling-Schulz M. 2007. Diagnostic real-
time PCR assay for the detection of emetic Bacillus cereus strains in foods and
recent food-borne outbreaks. Applied and Environmental Microbiology 73:
1892-1898
From C., Hormaz bal V., Hardy S.P. and Granum P.E. 2007. Cytotoxicity in Bacillus
mojavensis is abolished following loss of surfactin synthesis: Implications for
assessment of toxicity and food poisoning potential. International Journal of
Food Microbiology 117: 43-49
Granum P. E. 2002. Bacillus cereus and food poisoning. In Applications and systematics of
Bacillus and relatives, edited by Berkeley R., Heyndrickx M., Logan N. and De
Vos P. Blackwell publishing company. MA, USA. pp: 37-46
63
Granum P.E. 2007. Bacillus cereus. In: Food Microbiology: Fundaments and Frontiers, 3 rd
ED. Edited by Doyle M. and Beuchat L. 3rd Edition, ASM Press, Washington,
D.C. pp: 445-456
Griffiths M. W. and Schraft H. 2002. Bacillus cereus food poisoning. In Foodborne diseases,
edited by Clicer D. O. and Riemann H. P. Academic press, California USA,
pp: 261-270
Guinebretière M-H., Broussolle V. and Nguyen-The N. 2002. Enterotoxigenic profiles of
food-poisoning and food-borne Bacillus cereus strains. Journal of Clinical
Microbiology 40: 3053-3056
Guinebretière M-H. Fagerlund A. Granum PE. and Nguyen-The C. 2006. Rapid
discrimination of cytK-1 and cytK-2 genes in Bacillus cereus strains by a novel
duplex PCR system. FEMS Microbiology Letters 259: 74-80
Han, C.S., Xie, G., Challacombe, J.F., Altherr, M.R., Bhotika, S.S., Brown, N., Bruce, D.,
Campbell, C.S., Campbell, M.L., Chen, J., Chertkov, O., Cleland, C.,
Dimitrijevic, M., Doggett, N.A., Fawcett, J.J., Glavina, T., Goodwin,L.A., Green,
L.D., Hill, K.K., Hitchcock, P., Jackson, P.J., Keim, P., Kewalramani, A.R.,
Longmire, J., Lucas, S., Malfatti, S., McMurry, K., Meincke, L.J., Misra, M.,
Moseman, B.L., Mundt, M., Munk, A.C., Okinaka,R.T., Parson-Quintana,B.,
Reilly, L.P., Richardson, P., Robinson, D.L., Rubin,E., Saunders, E., Tapia, R.,
Tesmer, J.G., Thayer, N., Thompson ,L.S., Tice, H., Ticknor, L.O., Wills, P.L.,
Brettin, T.S. and Gilna, P. 2006. Pathogenomic sequence analysis of Bacillus
cereus and Bacillus thuringiensis isolates closely related to Bacillus anthracis.
Journal of Bacteriology 88: 3382-3390
Hallaksela A.-M., Väisänen O. and Salkinoja-Salonen M. 1991. Identification of Bacillus
species isolated from Picea abies by physiological tests, phage typing and
fatty acid analysis. Scandinavian Journal of Forest Research 6: 365-377.
64
Henriques A. O. and Moran C. P. 2007. Structure, assembly, and function of the spore surface
layers. Annual Reviews of Microbiology 61: 555-588.
Helgason, E., Økstad, O.A., Caugant, D. A., Johansen, H. A., Fouet, A., Mock, M., Hegna, I.
and KolstØ A. B. 2000. Bacillus anthracis, Bacillus cereus and Bacillus
thuringiensis - one species on the basis genetic evidence. Applied and
Environmental Microbiology 66: 2627-2630.
Holt J.G., Frieg N.R, Sneath P.H.A Staley S.T. and Williams S.T. 1994. Genus Bacillus. In
Bergey´s manual: Determinative bacteriology 9 th edition. Edited by Hensyl
W.R. Baltimore, USA Williams & Wilkins. pp: 559-564
Hoornstra D., Andersson M.A., Mikkola R. and Salkinoja-Salonen M.S. 2003. A new method
for in vitro detection of microbially produced mitochondrial toxins. Toxicology
in Vitro 17: 745-751
Hoornstra D., Dahlman O., Jääskeläinen E., Andersson M., Weber A., Aurela B., Lindell H.
and Salkinoja-Salonen M., 2006. Retention of Bacillus cereus and its toxin,
cereulide, in cellulosic fibres. Holzforschung 60: 648-652
Hormaz bal V., ∅yvin ∅., O´Sullivan K. and Granum P.E. 2004. Quantification of Bacillus
cereus emetic toxin (cereulide) in figs using LC/MS. Journal of Liquid
Chromatography and Related Techniques 27: 2531-2538.
Hornstra L. M., de Vries Y. P, de Vos W. M., Abee T. 2006. Influence of sporulation medium
composition on transcription of ger operons and the germination response of
spores of Bacillus cereus ATCC 14579. Applied and Environmental
Microbiology 72: 3746-3749.
65
Hornstra L.M, de Leeuw P.L., Moezelaar R., Wolbert E.J., de Vries Y.P., de Vos W.M. and
Abee T. 2007. Germination of Bacillus cereus spores adhered to stainless steel.
International Journal of Food Microbiology 116: 367-371
Horwood PF., Burgess GW. Oakey HJ. 2004. Evidence for non-ribosomal peptide synthetase
production of cereulide (the emetic toxin) in Bacillus cereus. FEMS
Microbiology Letters 236:319-24
Hoton FM., Andrup L., Swiecicka I., and Mahillon J. 2005 The cereulide genetic
determinants of emetic Bacillus cereus are plasmid-borne. Microbiology 151:
2121-2124
Hughes S., Bartholomew B., Hardy J.C. and Kramer J.M. 1988. Potential application of a
HEp-2 cell assay in the investigation of Bacillus cereus emetic-syndrome food
poisoning. FEMS Microbiology Letters 52:7-11
Häggblom M. M., Apetroaie C., Andersson M. A., and Salkinoja-Salonen M. 2002.
Quantitative analysis of cereulide, the emetic toxin of Bacillus cereus, produced
under different conditions. Applied and Environmental Microbiology 68 : 2479-
2483
IDF (International Dairy Federation) 1998. Dried Milk Products. Enumeration of Bacillus
cereus. Most probable number technique. Standard method 181.1998
International Dairy Federation, Brussels, Belgium.
ISO (International Organisation for Standardisation), 2004. Microbiology of food and animal
feeding stuffs - horizontal method for the enumeration of presumptive Bacillus
cereus-colony-count technique at 30°C. 3rd edition, Standard 7932:2004
Geneve, Switzerland
66
Ivanova N., Sorokin A., Endersin I., Galleron N., Candelon B., Kapatral V., Bhattacharyya
A., Reznik G., Mikhailova N., Lapidus A., D´Souza M., Walunas T., Grechkin
Y., Pusch G., Haselkorn R., Fonstein M., Ehrlich S. D., Overbeek R. and
Kyprides N. 2003. Genome sequence of Bacillus cereus and comparative
analysis with Bacillus anthracis. Nature 423: 87-91
Jay, J.M., Loessner M.J., Golden D.A. 2005. Bacillus cereus gastroenteritis. In: Modern food
microbiology, 7th edition, Springer Science+Business media, Inc., New York,
USA pp: 583-590
Jensen G.B., Hansen B.M., Eilenberg J and Mahillon J. 2003. The hidden lifestyles of
Bacillus cereus and relatives. Minireview in Environmental Microbiology 5:
631-640
Kawamura-Sato K., Hirama Y., Agata N., Ito H., Torii K., Takeno A., Hasegawa T.,
Shinomura Y., and Ohta M. 2005. Quantative analysis of cereulide, an emetic
toxin of Bacillus cereus. Microbiology Immunology 49:25-30
Kolst∅ A.-B, Lereclus D. and Mock M. 2002. Genome structure and evolution of the Bacillus
cereus group. Current Topics in Microbiology and Immunology 264:95-108
Kramer J.M. and Gilbert R.J. 1989. Bacillus cereus and other Bacillus species. In Foodborne
Bacterial Pathogens. ed. Doyle, M.P. Marcel Decker, New York, pp: 22-70
Lund T. and Granum P.E. 1996. Characterisation of a non-haemolytic enterotoxin complex
from Bacillus cereus isolated after a foodborne outbreak. FEMS Microbiology
Letters 141: 151-156
Lund T., De Buyser ML and Granum PE. 2000. A new cytotoxin from Bacillus cereus that
may cause necrotic enteritis. Molecular Microbiology 38: 254-261
Mahler H., Pasi A., Kramer J., Schulte P., Scoging A., Baer W., and Kraehenbuehl S. 1997.
Fulminant liver failure in association with the emetic toxin of Bacillus cereus.
New England Journal of Medicine 336:1143-1148
67
Mikami, T., Horikawa, T., Murakami, T., Matsumoto, T., Yamakawa, A., Murayama,
S. Katagiri, S., Shinagawa, K., Suzuki, M. 1994. An improved method for
detecting cytostatic toxin (emetic toxin) of Bacillus cereus and its application
to food samples. FEMS Microbiology Letters 119: 53-58
Mikkola R., Saris N.E., Grigoriev P.A., Andersson M.A. and Salkinoja-Salonen M.S.
1999. Ionophoretic properties and mitochondrial effects of cereulide: the emetic
toxin of B. cereus. European Journal of Biochemistry 263:112-117.
Moravek M., Buerk C., Broussolle V., Guinebretiere M-H., Granum P.E., Nguyen-The C.
and Märtlbauer E. 2006. Determination of the toxin potential of Bacillus cereus
isolates by quantitative enterotoxin analyses. FEMS Microbiology letters
257: 293-298.
Mortimer P.R. and McCann G. 1974. Food-poisoning episodes associated with Bacillus
cereus in fried rice. Lancet. 1(7865):1043-1045.
NCBI: http://www.ncbi.nlm.nih.gov/sites/entrez?db=genome&cmd=search&term=bacil
lus+cereus. 31.10.2007
Nishikawa Y., Kramer J. M, Hanaoka M. and Yasukawa A. 1996. Evaluation of serotyping,
biotyping, plasmid banding pattern analysis, and HEp-2 vacuolation factor assay
in the epidemiological investigation of Bacillus cereus emetic-syndrome food
poisoning. International Journal of Food Microbiology 31: 149-159
Niskanen T, Johansson T., Kuusi M., Raahenmaa M, Siitonen A., Tuominen P. 2006
Ruokamyrkytykset Suomessa vuonna 2005. Eviran julkaisuja 2/2006 sivut: 17-
27
Niskanen T., Johansson T., Siitonen A. and Kuusi M. 2007. Ruokamyrkytykset Suomesa
vuonna 2006. Eviran julkaisuja 21/2007 sivut: 15-28
68
Nordic Committee on Food Analysis (NMKL). 1997. Bacillus cereus. Determination in
foods. UDC 579.852.11. Method No. 67, 4th ed. Oslo, Norway
Paananen A., Mikkola R., Sareneva T., Matikainen S., Hess M., Andersson M., Julkunen I.,
Salkinoja-Salonen M.S. and Timonen T. 2002. Inhibition of human natural killer
cell activity by cereulide, an emetic toxin from Bacillus cereus. Clinical and
Experimental Immunology 129: 420-428
Parry J. M., Turnbull P.C.B and Gibson J.R. 1983. A color atlas of Bacillus species, Wolfe
Mecical Publications Ltd, London, UK, pp: 98-107
Pirhonen T., Andersson M., Jääskeläinen E., Salkinoja-Salonen M. Honkanen-Buzalski T.,
Johansson T. 2005. Biochemical and toxic diversity of Bacillus cereus in a pasta
and meat dish associated with a food poisoning. Food Microbiology 22: 87-91
Pirttijärvi, T. S., Andersson M.A., Scoging A.C., and Salkinoja-Salonen M.S. 1999.
Evaluation of methods for recognizing strains of the Bacillus cereus group with
food poisoning potential among industrial and environmental contaminants.
Systematic & Applied Microbiology. 22: 133-144
Pirttijärvi T. 2000. Contaminant aerobic sporeforming bacteria in the manufacturing
processes of food packaging board and food. Dissertationes Biocentri Viikki
Universitatis Helsingiensis 14/2000. Ph.D. thesis, University of Helsinki,
Finland
Pirttijärvi T.S.M., Andersson M.A. and Salkinoja-Salonen M.S. 2000. Properties of
Bacillus cereus and other bacilli contaminating biomaterial-based industrial
processes. International Journal of Food Microbiology 60: 231-239
Rajkovic A., Uyttendaele M., Deley W., Van Soom A., Rijsselaere T. and Debevere J. 2006a.
Dynamics of boar semen motility inhibition as a semi-quantitative measurement
69
of Bacillus cereus emetic toxin (cereulide). Journal of Microbiological Methods
65: 525-534
Rajkovic A., Uyttendaele M., Ombregt S.-A., Jääskeläinen E., Salkinoja-Salonen M. and
Debevere J. 2006b. Influence of type of food on the kinetics and overall
production of Bacillus cereus emetic toxin. Journal of Food Protection 69: 847-
852
Rasko D.A., Ravel J., ∅kstad O.A., Helgason E., Cer R.Z., Jiang L., Shores K.A., Fouts D.E.,
Tourasse N.J., Angiuoli S.V., Kolonay J., Nelson W.C., Kolst∅ A.B., Fraser
C.M and Read T.D. 2004. The genome sequence of Bacillus cereus ATCC
10987 reveals metabolic adaptations and a large plasmid related to Bacillus
anthracis pXO1. Nucleic Acids Res 32: 977-988
Rasko D.A., Rosovitz M.J,. Okstad O.A,. Fouts D.E., Jiang L., Cer R.Z., Kolsto A.B., Gill
S.R. and Ravel J. 2007. Complete sequence analysis of novel plasmids from
emetic and periodontal Bacillus cereus isolates reveals a common evolutionary
history among the B. cereus-group plasmids, including Bacillus anthracis
pXO1. Journal of Bacteriology 189:52-64
Reers M., Smiley S.T., Mottola-Hartshorn C., Chen A., Lin M. and Chen L. B. 1995.
Mitochondrial membrane potential monitored by JC.1. Methods in Enzymology
260: 406-417
Roberts T.A., Baird-Parker A. C. and Tompkin R. B. 1996. Bacillus cereus. In: Micro-
organisms in Foods. 5. Microbiological Specification of Food Pathogens.
Blackie Academic & Professional 1996 Great Britain. pp: 20-35
Rosenquist H., Smidt L., Andersen S., Jensen G. and Andrea Wilcks. 2005. Occurrence and
significance of Bacillus cereus and Bacillus thuringiensis in ready-to-eat food.
FEMS Microbiology Letters 250: 129-136
70
Sagripanti J-L., Carrera M., Insalaco J., Ziemski M., Rogers J. and Zandomeni R. 2006.
Virulent spores of Bacillus anthracis and other Bacillus species deposited on
solid surfaces have similar sensitivity to chemical decontaminants. Journal of
Applied Microbiology 102: 11-21
Sakurai N., Koike, K.A., Rie, Y. & Hayashi, H. 1994. The rice culture filtrate of Bacillus
cereus isolated from emetic-type food poisoning causes mithochondrial
swelling in a HEp-2 Cell. Microbiology Immunology 38: 277-343
Seibert H., Mörcher S., Gulden M. 2002. Factors influencing norminal effective
concentrations of chemical compounds in vitro: medium protein concentration.
Toxicology In Vitro 16: 289-297
Shaheen R., Andersson M.A., Apetroaie C., Schulz A., Ehling-Schulz M., Ollilainen V-M.
and Salkinoja-Salonen M.S. 2006. Potential of selected infant food formulas for
production of Bacillus cereus emetic toxin, cereulide. International Journal of
Food Microbiology 107: 287-294
Shinagawa, K., 1993. Serology and characterization of toxigenic Bacillus cereus. Netherland
Milk Dairy J., 47, 89-103
Shinagawa K., Konuma H., Sekita H. and Sugii S. 1995. Emesis of rhesus monkeys induced
by intragastric administration with the HEp-2 vacuolation factor, cereulide,
produced by Bacillus cereus. FEMS Microbiology Letters 130: 87-90
Shinagawa K., Ueno Y., Hu D., Ueda S. and Sugii S. 1996. Mouse lethal activity of a HEp-
2 vacuolation factor, cereulide, produced by Bacillus cereus isolated from
vomiting type food poisoning. Journal of Veterinary Medical Sciences 58: 1027-
1029
Sim R.B. 1998. Bacillus cereus gastroenteritis. In. Hausler W. and M. Sussman (eds.), Topley
& Wilson’s Microbiology and Microbial Infections, vol.3. Arnold, London,
Great Britain pp: 551-556
71
Svensson B., Monthan A., Shaheen R., Andersson M. A., Salkinoja-Salonen M and
Christiansson A. 2006. Occurrence of emetic toxin producing Bacillus cereus in
the dairy production chain. International Dairy Journal 16: 740-749
Swiecicka I. and Mahillon J. 2006. Diversity of commensal Bacillus cereus sensu lato
isolated from the common sow bug (Porcellio scaber, Isopoda). FEMS
Microbiology Letters 56: 132-140
Swiecicka I., Van der Auwera G. and Mahillon J. 2006. Hemolytic and nonhemolytic
enterotoxin genes are broadly distributed among Bacillus thuringiensis isolated
from wild animals. Microbial Ecology 52: 544-551
Szabo R. A. Speirs J. I. and Akhtar M. 1991. Cell culture detection and conditions for
production of a Bacillus cereus heat-stable toxin. Journal of Food Protection 54:
272-276
Taylor A.J. and Gilbert R.J. 1975. Bacillus cereus food poisoning: a provional serotyping
scheme. Journal of Medical Microbiology 8: 543-550
Teplova V., Jääskeläinen E., Salkinoja-Salonen M., Saris N.E., Serlachius M., Li F.Y. and
Andersson L.C. 2004. Differentiated Paju cells have increased resistance to toxic
effects of potassium ionophores. Acta Biochimica Polonica 51: 539-544
Teplova V.V., Mikkola R., Tonshin A. A., Saris N.-E.L. and Salkinoja-Salonen M.S. 2006.
The higher toxicity of cereulide relative to valinomycin is due to its higher
affinity for Potassium at physiological plasma concentration. Toxicology
Applied of Pharmacology 201:39-46
Thorsen L., Hansen B.M., Nielsen K.F., Hendriksen N.B,. Phipps R.K., and Budde B.B.
2006. Characterization of emetic Bacillus weihenstephanensis, a new cereulide-
producing bacterium. Applied & Environmental Microbiology 72 :5118-21
72
Turnbull P. C. B., Kramer J.M., Jörgensen K., Gilbert R.J, and Melling J. 1979. Properties
and production characteristics of vomiting, diarrhea and necrotizing toxins of
Bacillus cereus. The American Journal of Clinical Nutrition. 32: 219-228
van Der Zwet W.C., Parlevliet G.A., Savelkoul P.H., Stoof J., Kaiser A.M., Van Furth A.M,
and Vandenbroucke-Grauls C.M. 2000. Outbreak of Bacillus cereus infections
in a neonatal intensive care unit traced to balloons used in manual ventilation.
Journal of Clinical Microbiology 38: 4131-4136
Varnam A.H. and Evans M.G. 1991. Food poisoning: medical and microbiological overview;
Bacillus. In foodborne pathogens, an illustrated text. Mosby year book,
London, UK, pp: 9-19; 267-288
Vassileva M., Torii K., Oshimoto M., Okamoto A., Agata N., Yamada K., Hasegawa T. and
Ohta M. 2007. A new phylogenetic cluster of cereulide-producing Bacillus
cereus strains Journal of Clinical Microbiology 45: 1274-1277
Wang GYS, Kuramoto M., Yamada K., Yazakawa K. and Uemura D. 1995. Homocereulide,
an extremely potent cytotoxic depsipeptide from the marine bacterieum Bacillus
cereus. Chemistry Letters 9: 791-792
Warren S. C. and Gould G. W. 1968. Bacillus cereus spore germination: absolute requirement
for an amino acid. Biochemica et Biophysica acta. 170: 341-350
Wijnands L. M., Dufrenne J. B., Rombouts F. M., in´t Veld P. H. and Leusden F. M. 2006.
Prevalence of potentially pathogenic Bacillus cereus in food commodities in the
Netherlands. Journal of Food Protection 69: 2587-2597
Wijnands L. M., Dufrenne J.B., van Leusden F.M and Abee T. 2007. Germination of Bacillus
cereus spores is induced by germinants from differentiated Caco-2 cells, a
human cell line mimicking the epithelial cells of the small intestine. Applied
and Environmental Microbiology 73: 5052-5054
73
Yokoyama K., Ito M., Agata N., Isobe M., Shibayama K., Horii T. and Ohta M. 1999.
Pathological effect of synthetic cereulide, an emetic toxin of Bacillus cereus, is
reversible in mice. FEMS Immunology Microbiology 24: 115-120
Yousten A.A. 1975. Germination of Bacillus cereus endospores: a proposed role for heat
shock and nucleosides (review). Canadan Journal of Microbiology 21: 1192-
1197
74